Everything You Need to Know About Scaling eCommerce in 2026

In this video, you'll learn everything that you need to know about e-commerce in 2026. We spent over two hours on creative strategy, over an hour on offer design. We go into Google Ads technical account structure, meta ads technical account structure, as well as everything that you need to know about finance to be able to scale profitably. It doesn't matter if you're doing 7, 8, 9, or even 10 figures in revenue because almost all the content throughout this video is targeted at eight and nine figure brands. Now, before we start, if you're a performance marketer, please

reach out to us at hiringbluensedigital.com if you're looking for a role. And if you're a brand doing over $5 million a year in revenue, click the link in the description and reach out to us for a free audit where everything that you're about to learn in this video will get translated and applied specifically into your accounts to help you achieve scale. With that being said, let's dive in. I've completed over a,000 audits on seven, eight, and nine figure brands. And this bottleneck is the most common one that people completely overlook. It's data integrity and assurance.

What this really comes down to is using metrics that directly correlate to the actual commercial outcomes within the business and more using metrics that we can trust. So firstly, we're going to be going through why data integrity is the bottleneck in about 30% of businesses that I audit. Then we're going to go into platform reported rorowaz, why it's lying to you, how to fix it. We'll then talk about attribution windows, go into Better metrics, the best metric. We'll then show you how to reconcile the platform data against the P&L. We'll go into incrementality testing and

causal inference for those businesses that are a little bit bigger. Then we'll talk about tracking infrastructure, the nine patterns that we see in every Blue Sense audit so that you don't make the same mistakes. And then we'll wrap it all together through the playbook. The best way to explain why data integrity is a bottleneck in a lot of businesses is because every business is just a bunch of feedback loops. And so you collect data over here. You then make a decision with that data that then goes into action. Then you collect more data and you

make another decision. Flywheel effects. You get this specifically in creative within ecom. And so you will look at an ad down here and you will go this ad is performing well. It's at a 4x return. Amazing. We're going to do two things. Number one, we're going to go and make more ads like this. We'll tell the creative team. Number two, let's put way more spend in here. So you'll make the decision. Then you'll collect more data. Maybe rorowaz falls off a little bit and it goes to a 3x at the higher spend. And then we

go and launch a bunch more ads. and these ads are all performing well. Amazing. So, we reinforce into this decision even harder and we continue this feedback loop. Now, this entire thing breaks if this number right here was wrong. What if it actually wasn't at a 4X? But you made all these decisions under the assumption that it was driving a commercial outcome. But once again, what if it wasn't true? What if this ad actually wasn't driving top ofunnel awareness? What if this number was substantially overinflated? What if the number didn't matter at all? Well, then

all of a sudden you're making decisions within the business that seem to be making an impact when we zoom in on this one particular metric and this one number. But when we zoom out, profit isn't going up, revenue isn't going up, the business doesn't seem to be getting healthier, and we feel we're making decisions. We feel we're analyzing data and we're constantly iterating, but it's doing nothing. And that feeling of being on a treadmill and not moving is the feeling that your data is wrong. the metrics that you're using to make decisions are not actually

stepstone starting up or commercially aligning to correct decision- making that's fixing the business. And so when you feel like you're just spinning your wheels, it usually ends up being a metric problem. Now, I don't want to blame everything within an ecom business on just data integrity. Really, there's three pillars that we see in an e-commerce business, which is that there's the actual technical account structure. How are the different platforms being structured to be able to distribute budgets and learn based on the actual commercial objectives of the business? Number two is creative. What is the creative

velocity? What's the diversity? What's the strategy here? And then number three is data integrity and assurance and measurement. Now, even though I've put it in this order, data integrity actually forms the base. So then we can actually work on creative and then ultimately technical account structure sits up the top here. And the reason this is so critical is because the account structure and creative is meaningless unless the metrics actually make sense and we have congruent decision-making. Then technical account structure means nothing unless we have creative in place as well. That's it at a velocity that's

required. It's good creative and it's diverse. And then only does technical account structure matter if these two are actually in place in the first place. Now the reason this matters more than ever right now is because for the last 5 years, iOS 14, cookie depreciation, modeled conversions, and Meta's increased reluctance to actually expose the attribution mechanics have all pushed brands further away from causal truth. Platform rise has become a number that lives further and further and further away from the P&L. Most brands have responded by either ignoring the problem or they buy third party attribution

tools which just end up compounding the issue and causing even more confusion internally. The brands that are still scaling in 2026 are the ones that built a measurement layer that they can defend internally and that they can reconcile against actual commercial outcomes. And I cannot stress this enough. A business will always not grow due to one particular bottleneck. There's always one thing that is stopping growth. And then as you grow, another bottleneck appears. And so growing a business is the product of continuously solving bottlenecks over and over again. And that's what we do for clients

because ultimately our goal is to grow the business primarily through paid media. But if there is a bottleneck that is not allowing us to push paid, we will go and address it. Now the real question becomes okay if solving bottlenecks is the way to grow if that's really what you need to understand and be able to identify well then how do we identify bottlenecks as fast as possible know that that is the bottleneck that's where we should put all the resourcing and then continue to grow and the way that you diagnose and find bottlenecks is

through data and if the data doesn't have integrity if the data is wrong if the data is meaningless irrespective to our goals if the data is giving us false flags then we will never be able to diagnose the bottleneck. And so it all starts with having accurate, correct, measurable data that we can then use to go and identify where the bottleneck is in the business that we can unlock. If this isn't in place, data integrity ends up typically being the bottleneck in the Business. So platform reported rorowaz, which has been the gold standard for the

last 5 to 10 years. is I'm actually going into an RFP in a few days time with a 9 figureure brand and they are telling us how can we improve our reported rorowaz and the answer is we're not going to do that and I'm going to explain why and why this is a terrible KPI and if you are northstarring against this well you're probably 3 years behind competitors who are not operating under this model and they will ultimately beat you so you will ultimately have to move away from this kind of orientation into the platforms

but hopefully I can speed that process up for you by explaining why you shouldn't be doing this. So number one, all rorowaz uh is not equal and this applies through a multitude of different factors. So number one, this is all revenue isn't equal. So what that means is that we can have two separate campaigns. We can have campaign one over here and campaign two. They can both have a full return on ad spend on them and they could have both driven, let's say, $10,000 in attributed revenue. But the margin profile across these two different $10,000

of revenue could be very different because this one could be a discount campaign that has eroded margin. This one could have products that have a lower margin. This one up here could be a retention campaign that's full price whereas this one's using some kind of discount code on entry. The margin profile changes across different campaigns. And so rorowaz here and rorowaz here doesn't actually mean the same amount of profit contribution. And so we might go into a profit contribution calculation and find out that this top campaign is at three, This bottom campaign is at four.

And so if we actually look a little bit deeper into the profitability of these campaigns, we would say, "Oh, actually we want to spend way more money here. Even though it's the same rorowaz on the surface, this campaign generates a lot more profit." That is number one. Number two is that all rorowaz isn't equal across different platforms when we compare them. So if you take a really good example of this is Pinterest and you go and look at your Pinterest return on ad spend and then you go and compare this to we could give a

really extreme example here and we could call this TV, right? Cuz TV has a zero rorowaz on it because there's no attribution but let's keep it within the realm of digital and go meta. Maybe your meta rorowaz is a 2.5. Your Pinterest is a seven. Now you look at this objectively and you go well Pinterest is where we should put more money. Let's not put more money into meta. Let's go Pinterest. Issue is the way that Pinterest attributes is it includes viewthroughs. So if a user sees an ad and then buys, they don't have to

click on it. They don't have to interact with it. It will attribute the conversion. And then number two, if you've ever gone on Pinterest, there's a bunch of tiles all over the place. There's like six to 12 tiles depending on the size of your screen. As long as one of them is an ad, it can claim the view through conversion. And so what Pinterest tends to do is it will just reserve to warm audiences that have been on the website in the last few days. and that person was probably going to buy anyway. You get

to serve a pin to them. It gets to claim credit. You get massive overattribution. And so comparing these two numbers is a terrible idea. It's not a good exercise. Same thing with comparing this over to Google or another cuz they all have different attribution models. They all fundamentally work different and they all sit at different Stages of the funnel. So crossplatform rorowaz is not equal and all rorowaz is not equal across the funnel which is a development of this idea but it applies into the platform too. So if we have a meta campaign that's a

cold targeting campaign and the reason it's cold targeting is that we have existing customer lists excluded. So it is cold and we have very top ofunnel creative in here too. So we're also getting the algorithm to serve our ads to cold audiences because of the creative type. Then down the bottom here we have let's go to the absolute extreme and say that this is a bottom offunnel retargeting campaign for existing customers. Now this campaign down the bottom here might have a seven rorowaz and this campaign up the top here might have a two. Now naturally

once again you look at these two numbers and you go let's put spend here. This is a great return. This is going to grow the business. Let's not put it up here. But the issue is this sits at the bottom of the funnel. So the only way the business grows is if more spend gets injected at the top efficiently. If we start pulling spend away from the top and redirecting it to the bottom, that's when the business starts to collapse cuz we're not bringing new cold audiences in to begin to convert them. And you see

this all the time, particularly with something like a middle of funnel campaign that sits in here that's maybe at a 4x. You'll see all the time people will go, "Oh, let's just take 30% of spend and push it down here. Let's get some more spend in middle and bottom." You rarely ever need more spend in middle and bottom of funnel. Almost everyone overspends here. You need more spend on like a YouTube cold targeting campaign. You need more spend on Tik Tok. You need more spend on all of these really upperfunnel channels because this is what's

going to drive conversions down the line in these middle of funnel and Bottom of funnel campaigns. And I'll give you an example like a tricky example of how this will start to play out really confuse your decision-m which is that let's take this meta funnel right here and then let's say adjacent to this you decide as a business that you want to test out Tik Tok. So you go and launch Tik Tok over here and maybe you spend 25% of your meta budget on the platform doesn't seem to really do that well. It's at like

a 1x maybe a 2x. Nah, we can't really scale this. But at the exact same time, this middle ofunnel campaign here on Meta suddenly jumps from a 4x to a 12x. Now why did it jump? Well, it jumped up because all of this Tik Tok traffic that you were driving started getting retargeted on Meta and then started converting. However, most people will look at Meta and go, "Oh, our middle of funnel campaign is doing a lot better. Let's increase budgets and spend more here." They might even take it a step further and go, I think

that's probably because cold targeting is doing well. So, if we're going to up budgets here, let's up budgets on cold targeting, too, because overall meta is doing well. This had nothing to do with Meta doing well. Increasing the budget here is just going to be a waste of money. The reason why Meta is overattributing now is because of the Tik Tok spend. And so looking at rorowaz at a platform level, comparing platforms is not helpful. Looking at it through the funnel, it's going to change. And so all rorowaz is not equal. A rorowaz up here

of a two is much better than a rorowaz down here of a seven. And then this directional decision-m based on a return on ad spend number is going to mislead you in b budget allocations across the platform. Now there's three Main sources of overinflation in return on ad spend. Number one, which you'll hear me say so much throughout this video, is view through conversions. So, to explain this out of the gates on Meta, by default, you use 7-day click, one day view. And actually, these days, this will be 7-day click, one day view, one day

engaged. So, what this means is that if a user clicks on an ad and then buys within 7 days, Meta can claim the conversion, which is somewhat reasonable. If someone clicked and interacted with your ad and then purchased, we want that attributed to the campaign. One day view means if someone simply viewed an ad, they don't have to watch through the ad. They don't have to click on the ad. They don't have to interact with it in any way. It just needs to serve into their feed and then they have to buy within 24 hours.

Meta can also claim the conversion. And then lastly, one day engaged view is a definition of if someone watched more than 3 seconds or liked or commented or interacted but didn't actually do an outbound click. That will also get counted over here. Now, 7-day click is fine. And one day engaged is fine. But the one day view is where a lot of overattribution ends up occurring. Because if you're uh someone who's already visited the website, if you're an existing customer, you just get served an ad, you are going to buy anyway irrespective of whether the

ad served to you and then metagos and claims your conversion or attributes it to the ad. And so anytime you have one day view in your reporting, you end up with a rorowaz number that's way above the actual reality of the incremental impact of that campaign in the business. Number two is existing customer bleed over. So this is pretty obvious, but if existing customers are getting targeted Within your cold campaign. Well, you will end up with a bunch of overattribution because existing customers will view the ads. Existing customers are probably going to buy again anyway

and it gets credited in. Now let's say you removed user attribution from your attribution model. This will still overattribute because a click from an existing customer on an ad doesn't actually infer a causal relationship. And we'll go into causality in more detail a little bit later on. But the reality is that if I'm an existing customer, I've bought from you before. Let's say I'm buying supplements and then I get an ad from you for the supplement. Am I purchasing because I got the ad or am I purchasing because I'm like, "Oh yeah, I only have

5 days left in the pack and I might as well buy right now." And then I click through and buy. probably would have bought anyway regardless once the actual supplement ran out of my kitchen. It's just the ad reminding me a couple days earlier, oh yeah, let me get that through sooner. Is that actually genuinely incremental to the business? Probably not. And so you still end up with a good amount of over attribution here. And then lastly is the halo effect. And so you see this all the time with large retailers. If you're in a

big business or you've ran ad accounts for big brands, which is that they will do some kind of activation that sits way outside of the ad account, way outside of digital. Then everything in digital just improves. Okay, there's some kind of pop-up activation in New York and then suddenly all the performance in New York in the ad account triples and you're like, "Huh, if you didn't know that the popup and the activation was occurring, you would look at the ad Account and you go, I it doesn't make sense. The data makes no sense. How is

rorow so much better?" But it's because of this halo effect being caused from elsewhere. And so if you then made the decision, oh, our rorowes is up in New York this week. Let's pump spend. probably be a bad decision because the rorowaz going up actually has nothing to do with the causal impact of the spend, but it has to do with a third party confounding event that's then making it look causal when it's actually correlated. Now, it's worth noting here too that inflation will compound with the maturity of the business. So, this becomes a larger

and larger problem for the bigger the business is. And this is why this has become a very large focus of ours as we primarily now only work with 8, 9, and 10 figure businesses is that they pretty much all have this problem. This is why this entire video exists because data integrity and assurance is so critical in an 8 to 9 figure business, but almost no one is doing it actually well. Now, why does maturity cause a further overinflation? Well, it's because there's way more existing customers. So, there's way more opportunity for retargeting and overinflation.

There is way more channels that they're spending on. Now, this might even just be retail stores. You could think of retail stores as advertising because the store is there as people walk in. So, this also causes overinflation in online. And then a mature business that's been around for a while also likely has much larger organic presence. And so, the organic is overinflating all of the paid as well. So, the bigger you get, the less reliable return on ad spend becomes as a read at any level within the attribution funnel. So, how do we actually start

to improve return on ad spend then is there things that we can do? There is so much that you can do. What most teams will do Once they wrap their head around this and I'd say 50% of the market understands everything that I've just gone through. So, a lot of people are across this. And so, the solution becomes let's add let's add more. Okay? So, let's add in attribution tools. Let's add in more dashboards. Let's add in more conversion events. Let's add in more platform settings to try to fix this problem. Addition is not the

solution. We want to subtract. We want to remove as much noise as possible. We don't want to add more noise into the bucket. We don't want to add more metrics. We want to simplify everything down to where we have just core KPIs that are commercially orientated and we can make clear easy decisions. The more you add and add and add, the more confusion there becomes. That's when you sit in a meeting with a seauite and one person is going oh but the this attribution model on triple whale is saying that Pinterest is actually bad. And

then someone else is going, "Well, we ran an incrementality test on Pinterest and it was actually good." And then someone else is arguing that they ran a causal inference test and that it was actually bad. And there's all of these confounding voices based on all of this different data, which is why adding more data generally doesn't help. We want to strip data away. We want to get a priority, a hierarchy of metrics. What is the most important? That's the dec. If that's good, that's the decide. And then as you start to move down the hierarchy,

it gives an ability for you to prioritize what to actually do and what not to do. So in regards to attribution windows, I think it's really important for everyone to understand what attribution window your particular platform is using. So on Google, there Will typically be a 30-day click window, which means if someone clicks and then buys within 30 days, Google can claim the conversion plus typically a 7-day view, a 7-day engage view, and a 1-day view window. On Meta, typically by default, it's 7-day click 1-day view. Typically on Tik Tok, it's around about the same,

but you can change it, you can increase it, you can decrease it. Same thing on Pinterest, Microsoft's a little bit different, etc. So, firstly, you want to understand what is the attribution window on each of our platforms because that will change the incrementality factor of that number. Now, if we just silo into meta because that's where about 60 to 70% of our spend sits. By default, you will be 7day click, one day view, one day engaged. I recommend changing this on your campaigns moving forward to 7-day click, one day engaged. Remove the view throughs. Now,

you can even do this purely from a reporting perspective by hitting columns, compare attribution settings, and then opening up 7-day click reporting. This will correlate much more tightly to the actual efficiency numbers within the P&L, which we'll dive into a little bit later. The reason why we have one day engaged in here is because the 7-day click definition actually changed about 2 months ago from today. So, right now it's the 5th of May 2026. In I believe April, maybe March 2026. 7-day click used to be defined as someone who would click on an ad and

buy or someone who would interact with a post with a click. And so it didn't have to be an outbound click to the website. It could be a click on the like button. It could be a comment. It could be a share. It could be a click to the page. As long as it Was a click on the ad, and it didn't have to be an outbound click, it would count within the conversions, which I think is something that not many people knew. Okay? I could comment on an ad and then as long as I

bought, Meta would claim that as a conversion to that campaign on a click basis. Now, Meta rewrote the definition recently and removed it. And so now 7-day click has to be an outbound click. And so all of the 7-day click rorowaz numbers inherently got a little bit worse overnight because there was a bunch of conversions that were getting counted that don't get counted anymore. How they then factored in this is that all of those engagement clicks. So people that liked, comment, share, go to the page, this is now called a one day engaged. So this

is if someone engages. So we want to include this in here because this is the traditional definition of a click. It's just moved into the engaged audience. And if someone's engaging with an ad and then they buy, typically there's probably some kind of causal relationship. Now to take this one step further in meta, we also have something called incremental attribution which rolled out about 6 months ago now. Now you can look at your incremental attribution by once again going and hitting on columns, hit compare attribution setting, hit incremental attribution and break it out. Now this

is a really good measure because what it does is let's say you're targeting 10,000 people a day in this square. Meta will go and hold out 10% of users and measure if these people are still buying compared to people that are seeing ads. And let's say that of the people that are seeing your ads, 2% of them are purchasing from you. People that aren't seeing your ads, 1% of people are still Purchasing from you. Now, why is this the case? Well, typically this is the case in larger retail businesses because everyone already knows who you

are. People are already buying from you all the time. If you're if no one knows who you are, let's say you're doing $10,000 a month in revenue, obviously this will be 0%. People aren't going to buy from you unless they see your ad. But if people know who you are, they're going to buy from you regardless. A percentage will be buying from you. Now, we do a difference in difference calculation, which is just minusing the two. So 2 - 1 = 1. And we know that the actual impact of ads is not 2%, it's 1%.

So if this was saying a five return on ad spend over here, well, no, it's actually a 2.5 return on ad spend. So this is ultimately how uh lift tests work within meta. This is called a conversion lift test. Then what Meta does, and this is what most people don't actually know, is that Meta is not doing this in real time and then applying it to incremental attribution. So there aren't real-time hold out groups occurring in your ad account. Instead, Meta takes conversion lift experiments from your competitors and based on the reads that they get

of how incremental they are, it then goes and applies a factor into your ad account. And so the incremental attribution read isn't a accurate real lift experiment outcome within the ad account. It is a estimation based on your niche and industries conversion lift test that competitors have run. So it's not a perfect measure but it's good directional advice on what kind of campaigns are actually doing well using proper lift tests as the experimental data. Now before we wrap this section up I want to quickly mention attribution models which is using some kind of Third-party attribution

tool. So, why don't we just solve all of these problems by adding a third party attribution tool in that can track users using an external pixel who click on Meta and then click on Google and then let's say they go over to uh Pinterest and then ultimately they buy, right? We can just track this using a third party attribution tool. Why don't we do that? The reason being is that this is an unknowable reality. We can never track someone's conversion journey perfectly and so we just playing guessing games and adding more rather than subtracting. And

so we're introducing more confusion into decisionm. The real idea here is that the most valuable clicks within or the most valuable interactions within a consumer journey is the first click and the last click ultimately because we want to know how do people first find out about us that ultimately leads them to a conversion down the line and then how do they convert? Because then what we need is we need to crank [snorts] the top of funnel as much as possible on how people initially hear about us and then we need to make sure we modulate

the bottom of funnel enough to make sure that these people end up converting down the line. Now the tough thing with third party attribution models is the last click we pretty much always have accuracy on. We know when people are clicking and then buying, but we know that in the platform anyway. We don't need a third party attribution model to tell us where bottom of funnel is. It's pretty obvious. Well, what about first click? First click is almost unknowable. first click attribution doesn't exist because the cookies or the attribution always breaks at some point in

the line. And I say this because I've spent hundreds of hours in third party attribution tools trying to understand where the first click is occurring and It's like always inaccurate. You go into individual users who have purchased from you and you can look at their entire uh click journey across all the different platforms. Half the time it says that the first click over here came from brand search and you're like how is that even possible? How was the first time they ever interacted with the business from a brand search? Like obviously it wasn't obviously probably

something let's say it was well they probably saw their friend wearing the product back here and that was the first interaction with the brand. And so there was this virality component particularly in fashion with people wearing the product around and that's actually what caused the interest that then pushed to the brand search that then pushed through all of these attribution touch points. And so if we saw this we would go oh let's put more spend into brand search but that's not actually what we need. We need more of whatever this is over here. All right.

So what are better metrics? Well, you can think about how accurate a metric is correlated to the actual commercial outcome in the business based on these three layers of the pyramid. Down the bottom, you have attributed numbers. So, you have stuff like rorowaz, CPA, uh, multi-touch attribution, etc. Then, as we move up one, we look at finance grade level metrics. So, rather than relying on attribution, we're just looking at actual commercial outcomes. And so what is the actual profit contribution in the business? What is the me of the business? What is the acquisition me of

the business? Then we move all the way up the top here to the best causal read of what is actually occurring which is with studies which is where we take control groups and treatment groups and we make a change to the treatment group and we measure against the control to understand the actual causal impact of That change. Now, as you go up the pyramid, this becomes much longer and harder to do, and so the actual time to feedback slows down. As you move down the pyramid, it becomes a lot faster, but you lose reliability. So,

the highest reliability is up the top here. You ultimately want to be making very large decisions within the business based on lift studies or financial metrics. But if you want to make quick directional decisions dayto-day, the platform level metrics can be helpful as long as we understand the nuances of the attribution. So I want to start listing off the metrics in this section to give you an idea of what you can track, why I wouldn't track some of them and what they become helpful for. So the most common one you will always hear is me,

your marketing efficiency ratio. Now people calculate this in two different ways. They either take total ad spend and they divide by total revenue. And so this gives you a percentage. Or they do the opposite and they do revenue divided by ad spend which gives you an integer like a 5x. And so 20% me and 5x me this is the same number. We're just flipping the division over itself. Now the reason why these two different versions exist is this is more of a marketer metric and this is more of a CFO metric. The CFO wants to

understand what percentage of revenue is getting allocated to marketing spend so they can control it over time. Hence they want to see a percentage. A marketer wants to see ROI. If we spend a dollar, how many dollars are we getting out? And hence marketers will typically look at it in this direction. Now, one common mistake in how MEI is miscalculated is that people use total revenue rather than using net. And so, you always do want to delineate down into net revenue, which minuses out returns. Obviously, really important in fashion uh because if you look at

total revenue today, it will compress tomorrow once the returns get processed. And so, you always want to be going off in net value. Now, ME is a fantastic metric for CFOs because it allows you to have visibility into percentage allocation on the total P&L towards marketing and control it. As long as you can control this number and you can control gross margin and operating expenses, you can ensure consistent deliverability of a percentage profit target. Now, the issue with this metric is it's fundamentally a bad metric for marketers. So the CMO should not be held accountable

to this metric. The reason being is that paid media or at least most of your advertising dollars are not going towards driving returning customer revenue. Your advertising dollars are going towards driving new customer revenue because that is ultimately how the business grows. And this is a really important thing to understand which is that if you stop acquiring new customers, the business's revenue will look like this. And the reason it will look like this is because this is cohort growth over time. And so when people repeat in your cohort analysis, people will repeat at a high

percentage and then it will decline. And so maybe 8% of people come back after month one of buying the product, then 6% 4% 3% 2% 1% and then it asotopes down to quite a low number. If you stop filling up the bucket, your total revenue will just follow that curve. And so you need to Continuously increase new customer acquisition or be putting new customers in to be able to maintain revenue or increase it. And so we really want to make sure that the media spend that we're putting towards marketing is driving new customer acquisition. Therefore,

we want to KPI it against new customer acquisition. So instead, we have a me, which is acquisition marketing efficiency ratio. This is the exact same formula, but instead we're taking new customer rev and dividing by ad spend. And once again, this can be a percentage or an integer. It doesn't matter. And so this might be a 4x. Let me run you through an example of why this is so important. And I see this in literally every single audit I do, which is why KPIing on the right metric matters. So let's say in January you were

at a 4x me. Then in Feb, it jumps to a 5x. Now off the back of this, due to it being a bit of a lagging signal, the way that you're looking at it, you go, "Oh, actually Feb's really efficient. Let's spend more money in March." And so you go up and spend more money in March, and it compresses your ME back down, but you're fine because this is a target. Now, this all looks good. However, it doesn't take into consideration what caused this number to go up. Was it an improvement in efficiency in new

or was it just more returning customers? So, when you go and delineate down to acquisition me, what you might find out is that this was a two, this was then a 1.8, and then this was a 1.6. And so, what happened here was in Feb, you actually had an increase in returning customer revenue that propped up this number. New customer revenue Went down. But because you're using this number to make decisions, not this one, you go, "Okay, well, let's spend even more." And so you go and spend even more. It further erodess new customer acquisition.

This becomes incredibly unprofitable for the business, but you don't notice it because once again, you're indexing against me. And so you'll start to make decisions in terms of paid media allocation and the efficiency of paid against a metric me that isn't actually congruent to the efficiency of paid. Then a common question I get asked is, well, what is a good acquisition? M what should we be aiming for? And the answer to this is the same as the fact that all revenue is not equal. It's the same as acquisition me is not equal. And so in

some businesses with a 60% gross margin, you can obviously have a way worse acquisition me and still be just as profitable as someone else who only has a 40% gross margin. And so the target here is going to be respective towards the margin profile as well as the retention of the business. If you have incredible retention and you have like 50% lift by the six-month mark, then obviously you can have a much more aggressive acquisition strategy, make less money on first purchase because you know these people are going to come back and improve the profitability

of the business. So trying to compare AME against different businesses uh becomes quite difficult because they all have different margin profiles and different retention. The next metric that you have here, the finance grade metric is going to be new customer rev. Just tracking this independently. Most people don't do this. And then new customer orders as well. And so this is going to show us how volume is changing over time. And then we can look at efficiency separately. So this will tell us efficiency. This will tell us volume. Generally looking at CAC 2, it's similar to

acquisition me except CAC is going to show us what our cost to acquire is. The reason why you would measure both of these is that acquisition me is actually a better measure of profitability because it encapsulates average order value. CAC doesn't. But if our CAC fluctuates quite a bit, it's always good to be able to use that as a leading indicator to understand why is the product portfolio changing. What is actually occurring that's causing CAC to change irrespective of acquisition me? And for those that don't know, CAC is your total ad spend divided by the

amount of new customers acquired in that particular time period. Then as some final finance grade metrics here, and we could go all day on putting together a bunch of different finance grade metrics. There's literally tens of them. But the idea here is not to add, it is to subtract. You want the least amount of metrics possible to be able to have an indication on whether you're on or off track and whether to make uh changes off the back of it. And so the next one, the final one is going to be profit contribution. Profit contribution

is revenue minus cost of delivery minus marketing expenses. And this will give you a dollar value in contribution profit. You can also do this as a percentage. you would just min us the percentages here. Now, even better than this really is to delineate down into new customer profit contribution. And this is quite a self-explanatory. Rather than rev, you just do new customer rev minus the cost of delivery associated with this revenue minus the marketing associated as well. Now, this is going To tell us if we're profitable on acquisition or not. Really, you should have a

dashboard that allows you to track this number for the CFO. This number should be the headline metric at the top. These should be then the two headline metrics next to this. You then have CAC in the dashboard too. And then you have your profit contribution numbers. If you were tracking these numbers here, this is really all you need for a dashboard on where the business is right now, how profitable it is, whether the decisions we made in the last week were good or bad, and whether we should course correct. They have a super high degree

of accuracy because they are based on the actual financial metrics in the business. Now once you become a very very big business let's say high eight figures into nine figures then there becomes a lot more nuance in these metrics for example you cannot claim AM on meta spend let's say we're managing meta and there's a bunch of meta-pend if the business is also spending a million dollars a month on influences because the influences are going to drive rev and so we can't look at this 4x and go oh meta's doing amazing this month well we

spent a million dollars on influences that you're not actually encapsulating into this formula And so as more channels get introduced and as more complexity goes into the business, this is where even finance level metrics start to fall off in their reliability. And that is where we need to go all the way up to the top here and start to look at lift studies to understand okay how much is meta actually driving though as a proportion of this new customer rev. Now, at smaller revenue numbers, at smaller amounts of channels, you can just make the assumption

that all the spend drives, all the revenue, and you will have crystal clear decision- making and be able to scale very well, and you'll remove all the confusion. But as you start to get very big, that's where the top of the pyramid actually exists For. We then go from better metrics to the best metric. Now, the reason why I call this the best metric isn't necessarily because it should be at the top of your dashboard. It isn't necessarily because this is the only thing you should track. It's because it is comparable across time and it

is comparable across businesses. And that is what makes it a really good metric for being able to identify particularly across our entire client portfolio. Where's everyone sitting right now? Who's doing badly? Who's doing well? Who's doing incredible based on just one metric? If we used acquisition me, it's impossible to index all of our clients and understand who's doing well and who's not. Because for some businesses, a 4E is good. For others, it's terrible because of the gross margin profile, because of the spend on other platforms, like all of these different nuances. But LTGP to CAC

can be compared across anyone. I can ask one of our CPG brands, "What is your one-year LTGP to CAC?" And then I can go and ask one of our fashion brands, "What is your one-year LTGP to CAC?" And they're comparable. I can go, "Oh, that fashion brand's better." They are better on acquiring customers right now and they're a more efficient business. They probably, as long as they're controlling operating expenses, have a higher net profit margin. And so this becomes a very good metric for comparing against yourself historically and comparing against others. Now why is it

so good? Well, it's because anytime you have a ratio, it is generally a superior metric because you have a secondary pairing metric that allows you to contextualize the primary. If you just look at CAC, well, it has no context. If you just look at LTGP, well, it has no context. You bring them Together, the context is provided. Now, the important component here is well, what actual time value do you look over? If you're just looking at yourself, you can do whatever time you want. Okay? So, we could do 30-day LTP to CAC and that's fine.

If you start comparing against competitors, particularly in other industries, like if you compare CPG to fashion becomes very different because there's very different uh retention profiles. And so a CPG brand might operate on a very low first purchase GP to CAC compared to a fashion brand, but the CPG brand might make way more money because they have higher repeat rates over the following year. So you just want to extend this time period as you start to compare to other people. But if you're just comparing to yourself, 30-day, 90day, 180day, absolutely fine. Now what are the

benchmarks here? Benchmarks and let's call this benchmarks on 180day LTGPAC. If you are below a one, this is very bad. This is the equivalent of a six-month CAC payback period if you're below a one, which is not good. Um, unless you have a lot of funding, you don't I would never want to run a business or probably even work with a business with a CAC payback period greater than 6 months. Super high-risk position to be in. You're having to float so much capital to even operate this thing. I don't like it at all. 2 to

three. Well, let's go 1:2. 1:2 is an okay position to be in. This isn't bad. You're making money. you're profitable on acquisition, but you're not that efficient. And so, either your cost to acquire a customer is too high or you're not making enough in GP. Two to three, this is good. This is where you want to be. This is where you want to scale. Any brand that's in this range For us, we're looking at how we can increase spend. 3 plus, honestly, this is bad. And it's bad because you're leaving money on the table. If

you're at over a 3 plus LTGP to CAC on a 180day basis, you could be scaling way harder, capitalizing way more on the current acquisition channels that you have in place, but you aren't. Now, if this is due to a cash flow problem, then that's completely understandable and fix the cash flow. Don't take out large loans to be able to capitalize here. But ultimately, this is an unbelievable position to be in. Most people aren't in this position and therefore you should put the pedal down and actually capitalize on it. Now, just one practical reality for

everything we've been going through is it's all reliant on real-time gross margin numbers. The LTGP to CAC means we need visibility in gross margin. The profit contribution means we need visibility into gross margin. And so, what you need is that if you're on Shopify, make sure that you have cost of goods tagged up on all of the products because that is going to give you real-time gross margin visibility as well as how gross margin compresses during discount periods. This can then flow through into any external software into any Google sheet reporting etc. which is going

to give you an accurate visible view into gross margin. Without this you will have to make a percentage assumption which will be wrong and will substantially mislead you. Let me explain why. Let's say that you have a $100 product and it has a $50 cost of delivery on it. This means that you have $50 in gross margin, Aka 50% gross margin. And because of this, this is what you actually tell your agency or your internal team. You go, "Hey, we've done the math. We've looked at our reporting with the accountant over the course of the

last 3 months. And our average gross margin is 50%." Cool. We then go and bake that into all of our models. We take your revenue, we times by 50%. That's your profit contribution. That's your LTP to CAC. Everything is using this number. However, you go into Black Friday and let's say you have a 20% compression in pricing. So, you do a 20% off storewide. Now, your price goes down to $80. Now, yes, average order value might go up. So, overall in terms of basket size, you get more, but the units is what is going to

determine gross margin. So, the unit price goes down to 80. Your cost of delivery remains the same at 50. Therefore, your gross margin compresses to $30. This means your new gross margin is 30 divided by 80 which is 37.5%. So your gross margin compressed during this sales period by a lot. This is 12.5%. If we do a percentage of the decrease, this is like a 30% decrease from 50 to 37. So, it's really important that you have cogs tagged up or else all of the metrics that you have within your dashboards that you're measuring against

will be wrong because it won't encapsulate margin compression as you discount. So, now moving into incrementality and experiment design. Fundamentally, the problem with attribution is that it's correlation. And so, what we're doing is someone clicks on an ad, then they go and purchase, and we're saying that this click caused this purchase. But we don't Have proven causation. There's no experimental design that tells us that this click actually caused the purchase. We are just assuming, okay, someone clicked, then someone bought, probably had something to do with each other. Therefore, we should try to do more of

this and see if revenue goes up. Now, that's absolutely fine and it's pretty much causal. If you were a business that no one knows about doing 0 in revenue and the only marketing effort you had was Facebook ads, then yeah, well, if someone bought, there's probably no way they bought anywhere else than clicking on a Facebook ad. So, it's very tightly correlated. Although it's still correlation, it's good enough to be able to assume that it's causal. However, where the issue starts to arise is when you have multiple different places in which you capture revenue and

multiple different channels in which you advertise on. And so rather than advertising on just one platform, when you introduce a second platform, maybe a third platform. So let's call this Meta. Let's call this Google. Let's call this Tik Tok. And then we can capture revenue over here on the.com website. We can capture revenue here in the retail store. we can capture revenue here on Amazon or one of our wholesalers. Then correlation becomes very confounded and this is obviously the issue with attribution, right? Is that we start to spend more on Meta and how do we

actually measure this? How do we measure the revenue capture? Because it's going to drive revenue here, here, and here. And then this is likely going to drive more people over here, which is going to lead into here. So it starts to become very convoluted. And so the gold standard here is to run something that's Called a geo lift experiment. What this is is that you take a country and you split by states. Now, I'm not going to draw a country because my drawing is terrible. So, just imagine that this is a country and we're splitting

up by states. What we want to do is we want to look for which states closely match the other states in terms of multiple variables. Now, generally, you will do a clustering algorithm using historical ad spend. And so, we want to take states that have moved together historically in the ad spend distribution to them. We want new customer revenue to have been tightly correlated historically. And then ideally, we also want to throw in new customer orders because revenue can diverge from orders based on average order value fluctuations. So we use these three variables. We look

at all of the data in all of these states over time and we look at which of these states is tightly correlated with another one. Now the case usually is that there is no state that's tightly correlated with another state and so instead we start making combinations and the combinations have an even better effect which is that we can derisk out of one particular state and we can use multiple and so we might find and let's just use the US as an example here that we have Arizona let's have Texas and then over here let's

have New York and then California we might find that when you add Arizona and Texas's daily ad spend together it looks like this. And then when you add New York and California's together, it looks like this. So they move in the same way. Then we do the same thing for new customer revenue and it looks like this. Then we do the same thing for new customer orders and It looks like this. Amazing. Now we know we have two states that are tightly correlated and predictive of the other state. So at any point in time, we

can go back and we can look at Arizona and Texas. We can go what was the revenue here? Cool. We don't even have to look at California and New York. we can predict it with 99% accuracy. So we're looking to build a predictive model that tells us if we look at this and we don't even look over there, we know what's happening. Then that gives us the ability to set this experiment up and go okay well what we are then going to do this is what's called our intervention and during this day we're going to

go and introduce Tik Tok or let's say we're going to take our meta cold targeting campaign and we're going to double or triple the budget. Okay, so we go and let's say meta we 2x budgets. Then we still have our control states that aren't impacted. We have our treatment states and we want to see what happens. And what we might see is that it is still tightly correlated and then it grows. And so then we can go okay well what should it have done using the counterfactual. So the counterfactual is the control that's predicting what

the treatment would have done and we would have predicted that it would have done this. And so now using the prediction which is predicted using the counterfactual and then using the actual output we calculate the difference here. So how much additional revenue was captured and let's say it's $100,000. So there was an additional $100,000 above the baseline expectation. Let's say on Meta we spent an extra 40K in ad spend. Well, then we know that this $40,000 in ad spend drove this $100,000 lift. Now, it's a 2.5x on the dot, which is a 2.5 incremental rorowaz.

And this is ultimately the crux of a geo lift experiment. Now, the reason why this becomes so incredibly helpful is because you can have multiple different channels in here, too. So, we don't just need to look at revenue when we're looking at this graph. We could look at retail revenue. So what is retail revenue in this states versus this and we can see if there was an impact from metas-pend. We can then run this experiment and then when it finishes we choose different states and we run another experiment. So we can continue to measure the

impact on revenue realization. This also becomes critical when you go really high up the funnel. So if you were going to go and let's say do TV or out of home campaigns, this is a great way to measure it. You do out of home in just these states and you see is there a divergence from the control and if there is okay well now we can start to have a measurable outcome on our spend input on a really top ofunnel activity. Now there is a caveat here for anyone in Australia which is that you actually

can't do state-by-state design and incrementality testing. You can but it's it's really not great. You need massive spend levels and you can't run another one soon after because you muddy up the clustering data that's used to do state selection. And so if you're going to do geolo lift in Australia, you need to do it at a commuting zone level, which simply means you need to split the country up into even further sub regions so that you can have enough region selection to be able to actually run a Lift experiment effectively. Now this is a geolyft

experiment that you do outside the platform. You measure it against revenue in the business. There's also lift experiments in the platform. And so you can run a lift experiment in Meta, in Google, in Tik Tok. Usually they just need relatively large budgets. In Meta you can just run them all the time. Fortunately in Tik Tok they always give you a minimum spend. So you need to spend at least 200 grand to be able to actually run the test. Same thing with Google. Now the difference between a lift experiment in the platform and a geolyft experiment

is that because the platforms have each individual user ID rather than needing to split at a country level to segment the control and treatment they can just split at an individual user level. So I can go into bucket A and then someone else can go into bucket B. And then this person will see the ads. I won't see the ads. Then meta, Tik Tok, Google, they measure, do I still buy? They look at the pixel. They see does Nathan end up going to the website purchasing or he doesn't. But these people do. Cool. We can

do an incrementality calculation. Now the disadvantages of Lyft experiments in the platform is that number one it relies on all the data science and measurement of the platform and so we are trusting that Meta or Tik Tok or whoever it is is giving us a real accurate read out of the back of it. There's no way for us to test their math. We don't get any visibility into the actual raw data set. So we can't actually verify whether it is what they say it is. Now I'd hope that it's legit but we can't verify it.

And it's important to know that because in geolo lift experiments we can verify everything because it's all our data and we're running the experimental design. The second disadvantage of a lift experiment in platform is obviously the limitations of attribution. And so this is relying us on us being able to attribute these people purchasing to the website and the other people not purchasing to the website. And so Lyft experiments have disadvantages, pros and cons. Pros, they're fast, they're quick on meta, you can run them all the time. Cons are relies on attribution. We don't get any

visibility into the data. So it could potentially not be reliable. But geoloyft experiments too also have cons which is that they take a lot of time to run. They typically do require a lot of spend to be able to reach statistical relevancy in the spend increase required. Um and it relies on us having to control variables at least within the uh state selection through the test period. And so these aren't easy to run and these aren't just some magical solution and neither are these. They both have pros, they both have cons. You should have both

within your experiment design strategy. One last comment here in terms of another experimental design that uh probably only the nerds watching this video will pick this up and actually do anything with it, but that's using causal inference. Causal inference is an open-source package open source by Google about I'm going to say 11 or 12 years ago. And this is just a predictive model for using a counterfactual to predict what would have happened at the point of intervention. So given the Example before, imagine you didn't split the states. And imagine what you did was at some

point in time, let's call this new customer rev right here, new introduced YouTube advertising. And what you want to do now retrospectively is you want to go, okay, we introduced YouTube here. What was the impact on revenue? We think it made a difference, right? Revenue was slightly lower and now it's slightly higher after YouTube came in. But was that because of other factors? Would that have happened anyway? Was this seasonality? How do we know that truly YouTube caused this increase in revenue? And now without the geoloyft experiment you never can. Okay, we can never prove

causation here. But we can try to use a predictive model to be able to understand what it would have done in terms of revenue using third-party confounding variables. So your daily revenue, what causes revenue? Well, number one might be ad spend. Number two, there might be seasonality on particular keywords. And so we can use seasonality keyword data. Number three might be spend towards influences. Number four, inflation data might have some kind of correlation. Number five, and obviously you can go and list any kind of variable that you think is predictive of your revenue. Then you

load it into the open- source package called causal impact. I'm sorry, I put causal inference up here. This is meant to be causal impact. And then it will give you a predictive outcome of what this revenue would be using all of these metrics during this time period. And so ad spend will be doing something over time and it will draw the relationship. And then the seasonality of keywords will be doing something over time and it will draw a relationship and it will draw a relationship between all of these other variables to be able to predict

revenue. And so it might come out and go using all of these variables and how they've moved historically and then how they're moving now, your revenue would have done this. And now because of that, we can go, oh, so YouTube wasn't actually as effective as we thought it was. And this is the reality of most experiments is a lot of people have no idea how to set up experimental design correctly and so they get false outcomes all the time. Which is why this whole video exists and why data integrity is so important. People introduce YouTube

or Tik Tok or something and they'll do it in like October and then they'll have a great November and they'll go, "Yeah, YouTube was probably helping." How? Like how did you measure that? How do you know that YouTube drove an uplift during one of the best periods of the year? You don't. There's no measurement system. And then once you actually start to get some kind of uh causal experimental design here, you start to find out that all of your assumptions are wrong. And so the actual impact of YouTube would be this area right here. So

this is the predicted counterfactual, the red dotted line. And then this is the actual revenue. And so just like before, you might find that this additional revenue over this time period was let's say like $10,000. Ad spend was $10,000. So YouTube drives a 1x ROI in the business. Now, causal impact obviously isn't perfect because this is using just a predictive model using thirdparty variables. And so, is it helpful directionally and does it allow you to do some retrospective analysis? For sure. And we have it built into our own internal software so that we can do

things like this retrospectively. We can Go back and we can say, okay, you introduced Tik Tok back here. Let's do a causal impact analysis. And what do we think Tik Tok roughly drove? It gives us some direction. It gives us some idea. Okay, it seems Tik Tok did nothing. Okay, it seems Tik Tok did something. Um, it's a way to apply Beijian statistics into marketing, which is ultimately what we're always trying to do with measurement is we're trying to apply real data science, statistical modeling into any results that we're trying to draw. Then on tracking,

I want to make three quick comments and then move on. The reality is that we don't go too deep on tracking because brands that we work with are typically doing at least 10 $20 million a year in revenue. And so if tracking is not set up properly at $10 to $20 million in revenue, you have some big problems. I also don't want to go too instructional here on here's exactly how you do this because you can just YouTube for a dedicated video and find out exactly how to fix these things. Number one is Cappy. You

want to make sure that Copy is set up on Meta. This is your conversions API. How this works is that rather than purchase events directly firing from the pixel and sending back to Meta, the purchase event fires on your own local server and then it reroutes it back to Meta. The reason why that's important is that it removes the redundancy on the browser which has become super unreliable due to iOS ad blockers and privacy regulations. And it also enables much better event matching because it uses firstparty data. So it passes back the user who bought

their email address, their phone number, their customer ID. It obviously hashes all the data. So it's all privatized, but that way it can actually improve match rates. It also allows for offline conversion data. If someone buys in store and they use an email or a phone number, we could link it back to an actual ad being served on the platform. And so you want to make sure this this is actually working, it's going to improve your tracking, which is going to improve the amount of data flowing into the platform, which is going to improve optimization

on Shopify. You do not want to use any marketing or attribution reports in Shopify. They are generally quite unreliable. This is ultimately like the issue with attribution. Uh but Shopify in itself, I haven't seen much helpful data at all looking at the marketing reports in there and the attribution models that they're using. And then lastly, G4. G4 in my opinion is a redundancy layer. And so if everything was to go bad and our platform stopped tracking or something happened, we could lean back on J4 for some quick directional insight. We don't use J4 that much

at all for day-to-day decision-m because there's nothing in J4 that we just can't get out of Shopify or the platforms. And so at least as it stands in May 2026, I don't think G4 should really be a dashboard that you're opening to make decisions within the business on. I think there are much more important metrics that should be tracked to be able to index against the performance of the decisions that we're making. However, it is a good redundancy layer to just have there. So I would make sure everything is set up properly there. it's firing,

you've got your segmentation of campaigns, etc., so that if you ever do need to use it, if you ever do need to fall back, all the data has been collecting properly. Now, lastly, I want to run through the nine mistakes that we see in audits all of the time. And we'll fire through this relatively quickly. Number one is one Day view is running across all campaigns and is being used in reporting. And so there's massive overattribution. 30 to 50% of rorowaz is just coming from those views. And so once we remove it, we get a

much more realistic view of what is actually occurring in the business. Number two is the Google halo effect. Google's return on ad spend always looks way better than it actually is because it is taking credit for what is occurring on other platforms. Is that to say that you cannot acquire new customers through Google? No. Is that to say you can't go and spend $100 $200,000 a month on Google profitably that is genuinely net incremental? No. Of course you can. We have multiple clients that we are. But you have to take into consideration that Google will

overattribute. So when you see a 20 rorowaz on your pax campaign, even if it has brand excluded, you just need to be skeptical cuz you could just go and test put an extra $1,000 into that campaign. see if you get 20 grand back in topline, you probably won't. And so the incrementality of Google as a platform is usually overstated. And so you just need to be careful of the halo effect to be able to see and understand the real impact here. Number three is so much brand waste due to an over prioritization of rorowaz. And

so because rorowaz is the north star to get rorowaz up on campaigns, we just leave brand in. You wouldn't believe how many large 8 n figure retailers I see this on. And it just causes so much waste within the account. Number four is over segmentation. there's just way too many campaigns comparative to either the complexity of the business or the ad spend level that they're at. Now, this compounds into a measurement issue because you have all of this data fragmented and so decision-m becomes very convoluted. Number five is having a bunch of tertiary channels like

Snapchat, Pinterest, Reddit, all of These other third party channels which have such a low spend allocation and have so much over attribution because when spend is low, it will always just prioritize the warm audience that's hitting the website through the pixel and so you end up with incredibly high return on ad spend numbers on channels that just aren't driving any new incremental growth to the business. Number six is existing customer overspend which is just inflating all the numbers and they're increasing budgets here because rorowaz looks good. Number seven is top ofunnel objectives running on meta

like traffic campaigns or reach campaigns etc. Uh rorowaz is terrible. There's no rorowaz which is funny cuz you're going why would they run this if there's no rorowaz but they're running this because there's an assumption of performance but there's no measurement system in place to actually measure it. almost I I actually don't know anyone I've ever spoken to that has run traffic campaigns or is actively running traffic campaigns in their Meta account that has a way to measure it. Like how are we actually measuring the impact? Oh well, we started running them like a year

ago and our agency said revenue went up and so we just we drew like correlation that traffic campaigns equals better revenue. Okay, what was the test design like? Oh, we just like put a traffic campaign in at 3% of budget uh during Black Friday and Black Friday was pretty good. Okay, well that 3% of budget is probably gonna have no impact on the business. Number one, no measurable impact, particularly in just a before and after like causal impact trade. But number two, now we're just wasting 3% of budget on something that we can't measure. And

so if you're going to run something like a traffic campaign or a reach campaign, that's fine. But have a measurement system in place to actually be able to validate it. Otherwise, there's no point in running stuff that you don't know whether it's working or not. You're just making assumptions. Number eight is overspend on DPAs. This is a function of rorowaz looking really good. And so people go and spend more and more and more, but it's not actually incremental because it sits a bottom of funnel. Number nine is no actual testing structure in place, which really

plays into number seven here, which is that if you are going to run something, you need to have a testing or an experimental design in place to be able to validate whether it actually works or not. Because if you can't validate if something's working, it's not even worth doing. Like what is the point if we don't know if it actually is driving any kind of outcome to the business? So what actual action items should you take right now dependent on your revenue level? 1 to 5 million. You should switch over to 7-day click reporting on

Meta. It's going to much better correlate to acquisition me in the business. You should tag up Shopify cost of goods on all your products so you can get real-time visibility into gross margin and how that's going to impact the next metrics which is have a dashboard set up that gives you a daily read on acquisition me profit contribution, new customer profit contribution, new customer revenue, new customer orders, and blended CAC. And then lastly, make sure that you're excluding existing customers for topfunnel campaigns and that you're minimizing brand search spend at 5 to 20 mil. You

want all of this, but you want to start layering in lift studies within the actual platforms, top offunnel experimentation that's being measured against those lift studies and using incremental attribution for a second validation layer beyond 7-day click. Then once you start to go to $20 million plus, you want to be running geolo lift tests outside of the platform. You'll have complexity now in the channel mix as well as the revenue capture mix here. And so you want to be validating, you want a dashboard that ideally pulls in those omni channel metrics from any other revenue

capture place. And you can Actually go and add one or two additional metrics on top of this baseline dashboard once you get to this scale. And then you want incrementality factors in here, which is actually the metrics that I would recommend adding. So when you run a geoloyft experiment and you get an incremental return output, you use this as a factor that's applied to the inplatform attributed row numbers and then that will give you a more realistic view of how incremental the rorowaz numbers are across all your campaigns in all your platforms. The only way

to do this though is you need to run consistent geolyft tests a lot in a very high frequency so that you can get incrementality factors across all the channels and all the different campaign types that you can then layer into your reporting. Most brands that fail to scale paid media don't usually actually have a paid media problem. They have a measurement problem that impacts their decision-making loops. They're optimizing to numbers that don't even match what is actually occurring within the business. And the numbers that they are reading to be able to understand what is doing

well and what isn't is actually misconstrued or misunderstood as to where it sits in the funnel and how the attribution model is actually working. The brands that are winning in 2026 and 2027 are the ones that understand all of this and are moving towards financial based KPIs, incrementality tests at the larger levels of revenue or just cleaning up reporting in platform if that's what's required. If you're a performance marketer and you've gotten this far, please reach out to us at hiringbluense.com.au. We are always hiring for for A+ performers within performance marketing. And if you're a

brand that's gotten this far, feel free to reach out to us for a free audit as long as you're doing at least $5 million a year in revenue. Uh we will put together an audit for you that will be structured through technical account structure, creative and data integrity and assurance. So everything here will be able to translate into the actual data that you have. We'll be able to show you use cases of where you might have been optimizing for me, but Ame would have been better. And uh we'll give you the dashboard and reporting that's

going to facilitate this whole thing. And if you're not a PM or you're not a brand, subscribe. Most e-commerce brands are looking at the wrong number. They look at rorowaz, they look at revenue, they look at me, but very few of them actually understand how cash moves inside an e-commerce business, what financial models to use, and how all of this ties directly into paid media. And that gap between what the ad account says and what the bank account says is where most brands get into trouble. Today, I'm going to walk you through every financial concept

that you need to understand. This isn't theory. This isn't MBA stuff. This is core application that we use every single day with the eight and nine figure brands that we work with. Finance knowledge ends up being the bottleneck for a lot of agencies and a lot of founders because the ad account isn't the business. Platform metrics end up being a proxy for performance. Most teams end up misdiagnosing their problem because they think they have an ads problem or a media buying problem, but a lot of the time it's a measurement issue or it's a margin

issue or it's a cash flow problem. Good paid media strategy should always reconcile directly to the P&L. If the ad account looks good, but the business isn't improving, something is fundamentally broken. Better data should lead to better decisions and a lot of reporting done internally in large 9-f figureure businesses to seues as well as Small six to seven figure businesses between an agency partnership. A lot of it is vanity. So, we're going to strip it away in this video. Firstly, we're going to cover the P&L. We're going to be going end to end on every

component of the P&L and how it translates into direct to consumer e-commerce and changes the decisions that you make on a day-to-day basis. Then we're going to go into unit economics. We're going to break down everything in unit economics that is important for day-to-day decision-making as it ties into paid media and the business as a whole. We're then going to be touching on metrics. This is the metrics that actually matter. I'm not going to be giving you 20 different metrics that aren't actually going to move the needle or change your decisions. You want metrics that

fundamentally change your behavior after you read them. We then have cash flow verse profit. This is something that most smaller founders don't understand well enough. I guarantee every agency that's either watching this or an eight or nine figure marketing manager that's working with an agency. The agency doesn't understand how to properly integrate cash flow versus profit decision-making into the ad strategy. And so we're going to be breaking down exactly how you need to be doing this, where the complexity of cash flow gets introduced into an e-commerce business versus any other type of client that you're

working with across paid ads. And then lastly, we're going to go and put it all together. We're going to loop the P&L into unit economics, into metrics, into cash flow versus profit so that you can make better decisions and you can know everything that you need within finance for ecom. Starting off with the P&L, we need to start at the very top of the P&L, which is revenue. Now, there's two really important components to understand here with revenue. Number one, there is a difference between gross revenue and net revenue. And the differences in Decision-making are

enormous between them. So, we're going to break that down. Number two is a little bit more simple. A lot of people know this, but people don't think through it because they don't run a 10, 20, $30 million business, which is that revenue is not profit. And so, you can have a $10 million business that's running at 5% net profit and you'll be making 500k per year in profit or income. Now, technically, this won't actually flow through to income in an e-commerce business cuz net profit does not equal cash flow. Or you could have a $5

million business that's at a 20% net margin, which is healthy in ecom. And that means that you're making $1 million in profit. I would much rather every single day of the week own this business over this one. And the reason being is that a $10 million business is a much more stressful asset to hold than a $5 million business. And this is doing double the profit. I think in agency land and in revenue land, a lot of people overprioritize just arbitrary revenue growth with the idea that margin will expand. For a lot of early stage

founders, margin actually never ends up expanding and they just get themselves into a bigger and bigger and bigger business with the exact same amount of profit that they were making 3 to four years ago. And let me quickly expand on that statement. What do I mean by people scale to expand margin, but it doesn't actually work? Well, the logic when you scale an e-commerce brand is that you have three buckets of expenses. You have cost of delivery. You then have marketing. And we're going to break all of this down and I'll show you exactly what's

in each of them. And then you have operating expenses. Now, operating expenses should stay relatively fixed in an e-commerce brand because you don't need to expand people that heavily. So, as you scale up, what should happen is operating expenses should remain stable. So, the blue line here will be opex. Marketing will go up as revenue expands. Cost of delivery will go up as well. And then here's revenue at the top here. So when we then calculate out profit, what's actually happening to profit? It's going up over time. It's actually expanding as a percentage on the

P&L because operating expenses are remaining flat. Now the issue is is that for pretty much every small e-commerce founder that I've worked with, and for context, in the first four years of the agency, we worked with nearly 300 7 figureure e-commerce businesses. So I have a lot of experience working across that size of business. And what almost always happens is that seven figure founders are just not good capital allocators. And so operating expenses actually ends up going up faster than me and cost of delivery and therefore profit remains flat. It remains stable as a dollar

figure which is obviously not a position that you want to be in. Now let's get back to revenue and breaking down the P&L. When it comes to revenue, you have net revenue and you also have gross revenue. Now the difference here is that net revenue accounts for returns, refunds, chargebacks, sometimes a discount allowance as well, whereas gross is simply their cash collected on topline. Now you actually see this within Shopify, but in other platforms you won't. It's also important to note that the Shopify gross to net breakdown typically doesn't flow through into accounting softwares like

Zero, like QuickBooks. And so you will only get a net or a cash read within a P&L accounting software. You won't get a gross read. And this becomes really Critical for using the P&L to actually make decisions. So when we look at gross to net, this is really the data that you want to be looking at. And the reason being is that if you just have net revenue at the bottom here and you're looking at this within your accounting software, let's say you have 100K one month and then you have 110K the next month. When

you look at these two revenue numbers, there's not really much that you can substantiate out of it. You can go, okay, revenue increased. We don't really know why. We need now need to look elsewhere in the P&L. Let's go and look at marketing. Did marketing expenses increase? Let's go and look at operating expenses. Did we add another staff member that's driving some kind of revenue in the business? We have to start looking elsewhere. But the answer could actually be above this net revenue number. If we go and look above net revenue, what we might actually

see is that gross revenue across these two months is the same. Maybe it's 120K here and 120K here. But when we then go down to refunds, refunds in this month were at 20K, but refunds in this month were only at 10. Then we go down to discounts. There was zero change in discounts and maybe shipping charges. All shipping is free on this store. And so then when we look at the difference in revenue for this month, the difference in revenue is actually coming from refunds, which is a line item that doesn't get encapsulated in most

accounting softwares if you're not actually pulling through the gross to net revenue dynamic into the software. So really, really important that anytime you're looking at revenue, we're not just looking at net, but we're looking at gross through to net to understand these different levers cuz I could give a hundred different examples of this. Okay, we could change the discount line item. So actually refunds were stable across both months but the core difference was that there was actually 10k in discounts in the prior month and so because we had some kind of discount running or discount

code that didn't get encapsulated in the following month and therefore that's why we saw revenue expansion. These are fundamentally levers that exist within the business to increase profitability, increase net revenue. If we can decrease refunds, we make more money. If we can decrease discounts, we make more money. If we can increase the amount of shipping collected at checkout, we also make more money. And so you want to be looking at all three of these levers because often there actually is profit to be unlocked in better optimizing these three numbers. If you can push refunds down

a little bit, if you can push discounts down a little bit, if you can collect a little bit more shipping at checkout, which most people can do, most people underolct at uh checkout, you can improve the P&L. Next layer of the P&L is cost of goods sold. This is where people make a massive mistake in direct to consumer ecom, which is that they don't encapsulate all the actual expenses associated with cost of goods sold. Cost of goods sold isn't just the cost that you paid to the manufacturer for the item, but it's the total landed

cost of getting that product from the manufacturer to yourself, then into the customer's hands. And so we have the product cost. We then have importing taxes and duties. We then have the landed cost. So this encapsulates the freight to get it to your warehouse. And then sometimes included in here, sometimes categorized into shipping and fulfillment instead is the cost of shipping to the actual customer. So this would be shipping charges. And this will also encapsulate uh import taxes and duties if you're shipping internationally. Tariffs in as well as an example. So sometimes you'll go and

encapsulate that here and just count it in cogs. Sometimes you'll separate the definition to shipping and fulfillment. But overall both of these sit under a category called cost of delivery. And so whenever COOD is used or cost of delivery, it is encapsulating all of the cost of goods expenses and then all of the shipping and fulfillment expenses. And this is where people really mess up product margin versus gross margin. This definition is super misunderstood. When you say these numbers to people, you'll always get different answers. So let me run through an actual worked example. Let's

say that you're selling a $100 t-shirt. What we will typically get from a brand is they'll say, "Yep, on this t-shirt or on this category, we have 70% gross margin." But what they're typically doing here is they're going to someone internal within the wider team if it's a 8 n figure retailer or if it's a smaller business, they're just going and looking at their PO and they're going, "What did we actually pay for this t-shirt to the manufacturer?" And we paid $30 per unit. And therefore, they're going, "Well, 100us 30 is 70. 70 divided by

100 equals 70%." Now, this is product margin. This is not gross margin. And why this is so critical is because this GM number will be used in setting KPI. If we know what the gross margin is, we therefore know what percentage allocation we should be putting towards marketing to be able to ensure that we sell through this product. Now, with 70% gross margin, we would honestly probably be fine spending up to 20, even 25% on marketing to drive sales here. However, if this is actually a much lower number and we don't know about it cuz

we're getting given product margin rather than gross, well, all of a sudden we're driving this product at an unprofitable marketing efficiency. So, We then actually find out that it's $11 on average to ship this to a customer. It's then $4 in pick and pack fees at the warehouse. There's a 3% transaction fee here, which is actually something I missed in the cost of delivery expenses before, which is transaction fees. They always need to be accounted for in cost of delivery because it is a cost to fulfill the product. You have to pay that transaction fee.

A lot of people put transaction fees in operating expenses which is a mistake because the expense is variable with revenue. And so we have $3 on transaction fees. We actually also have tax implications here. So of this $100, we were actually including tax which was $9, which is a flowthrough expense. So we need to remove that. We then have the actual cost of landing this product to our warehouse. So yes, it was $30 a unit, but the actual PO cost a few thousand to get to us and ship via air freight. And so if we

then cut that down into a unit by unit cost, that's about an additional $4 per unit. And then the last thing here that we still haven't taken into the picture is an assumed refund or discount allowance. And this is absolutely critical in the fashion niche because in fashion refund rates can range from 10% in some cases I've seen up to 80%. And so gross margin substantially changes when 80% of the product is getting returned. And so we also need to factor in a refund or chargeback allowance. And then let's also throw in a discount allowance

too because people might be using uh coupon codes here at some particular rate. And so let's have $15 of allowance here because this allows for about $5 in discounts to be used on average as well as about a 10% refund rate which is Pretty standard within fashion. Now we go and add all these expenses up and we've got $46 in additional expenses to actually deliver on this product. That comes out to $76 in cost of delivery. So if we recalculate our gross margin here, which you do by taking price, you minus the costs and then

you divide by the price, this equals 24%. Now, this is an incredibly extreme example where all of the additional variable costs here have significantly added out outweighed the unit cost. But you can see how there can be such a drastic difference between the assumed gross margin, which is product margin, and the actual gross margin within e-commerce. And it's because in e-commerce, you have all of these associated additional variable cost that aren't there necessarily in brick and mortar retail. So, as we move down the P&L design here, we've gone through revenue, we've gone through cost of

delivery, gross margin. Now, we're at marketing. Marketing is relatively straightforward, which is anything associated with marketing and advertising the product falls into this bucket right here. So, this is anything on paid ads. This is any influencer payments, this is any events, this is any out of home. The big question here becomes does content production go into this bucket? And this is based on whether it is a variable expense or whether it is fixed within the business to meet the content demand. And so what typically happens through an e-commerce direct to consumer business is that content

velocity or volume needs to flex throughout the year in conjunction with revenue expectations. And so a typical e-commerce brand will look flat through Q1. We'll have a spike at June for end of financial year sales. Will then build back up, have a big November, December, and then fall back off. And so your creative production or amount of content entering into the ad account as well as obviously investment in influences, events, etc. needs to map to this. When you try to map creative to this flexing in demand, it becomes very difficult to do with just an

in-house team because if you have four people, they're probably underworked here. They're probably substantially overworked here. And then this is probably the only time a year in which they have the right amount of work. And so because of that, it's not really an ideal model. So, what you do instead is you have in-house here and then all of this additional flex up in creative volume requirements is typically done by agencies or you could pull in short-term hires that are just there to fulfill a 3-month period. Now, these agency expenses that flex, this is a variable

expense that should go into your ME. These fixed costs down here, this should be associated with OPEX. This is a fixed expense that needs to be held for indefinitely into the future while content is king for being able to drive revenue. Now, as you then move into flexing this throughout the year, that's when you can make a variable. Now, some people would also argue that, hey, let's just put all of this in the me bucket because all of this is revenue driving, which I also understand, but that's going to be where there's a little bit

of custom design within how you're approaching the P&L at an individual brand level. Once we now go past ME, which is the second big expense bucket within an ecom brand, the biggest expense is typically cost of delivery. Second biggest expense is me. We now get to another margin, which is contribution margin. This is one of the best metrics for indexing performance over time. And I'll tell you why in a second. Now, contribution margin is relatively straightforward looking at this P&L design. It is total revenue minus cost of delivery minus all marketing expenses. And this will

give you your contribution margin. Now, this can all be calculated as both a percentage and a dollar figure. So, revenue is 100%, you have cost of delivery at 30%, me at 20%, therefore your contribution margin is 50%. Now, why is this contribution margin number so helpful? Well, it's because if we're looking at it as a percentage, we are one step away from net profit. To get to net profit, all we need to do over here is minus off OPEX. So if we know that opex in the business is kept relatively stable at let's say 10%.

Well then that means that 50% minus 10% is 40% net. Now we will have a net target as a business. You might want to be holding 10% net margin or 20% or 30% net margin. You can then back math that into a contribution margin target. Right? So let me give you an example. Let's say you have a 20% net profit target within the business. Great. If you know your operating expenses will always sit at 15%. We then just need to work our way up the P&L. So over here we're starting at the bottom and we're

working our way up and we can go okay net plus opex means our contribution margin needs to hit 35%. So we effectively have a contribution margin target that's really far up the P&L that ensures that we hit a net profit target. Same thing with dollar values too, right? So we might have a net profit target which is a million. We might have a million dollar in operating expenses. And so our contribution margin target is these two added together, which is $2 million. And so we know for the year we need to produce $2 million in

contribution margin to be able to get a million in net profit. Now, one further piece of clarification here is that you'll actually hear contribution margin being used, not in this definition. So technically, this definition is what's called contribution margin 3. Contribution margin one is product margin, which is what we talked about at the start. Now why people call this contribution margin one I have no idea. People are just making it complicated. Okay you can just call this product margin but sometimes people call this CM1. CM2 is gross margin. So we also went through that at

the start. This is including all variable expenses associated with cost of delivery. And then CM3 is exactly what we just went through here. This is typically called contribution margin. Just really important delineation because sometimes people will say contribution margin and they'll be referencing one of these other definitions which is a weird thing to do. I don't know why people do it, but just make sure you understand those definitions. The third largest expense bucket is operating expenses. Now, how to think through what actually sits in operating expenses is very simple. It's just anything that doesn't go

in cost of delivery or marketing. Now, the typical three big expense buckets here within direct to consumer is going to be people at number one. It's going to be software at number two, and then it's going to be Some kind of office or fulfillment center here. Now it's important to delineate that actual product fulfillment like a 3PL or a warehouse the objective is to fulfill actually sits in cost of delivery. Same thing technically with warehouse staff. So if you have staff that are pickp packing and shipping product, you actually technically want that defined in cost

of delivery. What you do is you take their salary and let's say take their salary per day. So maybe it's like $400 per day and then you divide by how many packages they ship per day. Let's say they manage 100. Well, then you would take a $4 per order expense and you would move that over to your cost of delivery in your actual accounting software. You can tag particular staff and your bookkeeper can do this and then they can reconcile into a custom P&L report that will have warehousing staff in your cost of delivery. So,

you'll have an accurate gross margin number on a month-by-month basis that encapsulates people that are actually fulfilling the orders. And then who you would want in operating expenses for people is everyone that's not associated directly with the actual fulfillment of orders. So this would be like your head of marketing, your head of operations. List goes on. Now the real key to operating expenses and where most seven figureure e-commerce brands actually get this very very wrong is that they overinflate operating expenses believing that this is the way to grow the business because you commonly hear reinvest

into the business. That's how you grow. But in e-commerce the way that you reinvest into the business is actually to invest in marketing. This is the primary growth lever that exists within the P&L for a business like this. Going and simply hiring more people generally isn't a revenue generating exercise. Going and simply getting more software generally isn't a revenue Driving exercise. Getting a bigger office generally isn't a revenue driving exercise. None of these expenses generally drive much revenue. Now people are by far the greatest leverage that exists in any business. In fact, the ceiling of

revenue within a business is generally the ceiling of the summation of skills and talent within the people that exist within that business. So, I'm very much so of the opinion that there is nothing more important on this entire P&L than the people line item. However, seven figure e-commerce brands generally don't have the skill set to hire incredible people yet. Number one, they don't have a big enough business to attract incredible talent. Number two, they simply haven't hired, trained, and gone through enough interviews to be able to identify what those people look like. And I say

that from personal experience. I've personally hired over 60 people in the last 5 years. And at the start, I had no idea what to look for. I was just going through interviews trying to figure it out. And over time, as you do more and more interviews, and to date, we've probably done near 500 to a,000 interviews. We have a pretty good idea of what a really good high performer looks like verse not. But when you're getting started, you don't. And so people overinflate their people expense. They overinflate software because they think this is going to

drive revenue. They end up overinflating office and tertiary expenses and it ends up destroying the P&L. So this is just a really important one to watch because honestly this is the in a larger business this is the CFO. In a smaller business this is the owner's responsibility to be able to keep under control. So we can now run through the actual waterfall of the P&L here. So starting at the top and I'm going to use small numbers to keep this simple. we have $100 in revenue and let's just think about this as a daily revenue

on a tiny business so it makes sense. Then we Move down to cost of delivery which is $40 and so this business ends up with a $60 gross margin. Then the actual marketing expense here is 20% of revenue. They maintain a 20% me which means that contribution margin is now $40. Operating expenses these guys are keeping it at 20% as well which is $20. meaning all the way down the bottom here, we get to $20 in what we can define as net profit. Now, what you'll notice here is that I'm calling it net profit here,

but over here it's called Ibida. The reason for that is that technically this $20 here isn't actually net profit. Net profit sits after we take out interest, tax, depreciation, and amotization. Now, a lot of words if you're not familiar with finance, it's not really something that we need to get too deep into in this video. The only caveat that's really worth noting is that typically to fund an e-commerce business, most people rely financial tooling like loans. And the reason for that is it's very capital intensive. To actually be able to grow quickly, you need capital

to make future inventory purchasing whilst also continuing to spend on media. And now with that, you have interest repayments on those loans. Question becomes where does interest repayments go? A common mistake I've seen on a lot of seven figure and actually eight figure P&Ls is that interest expenses will go into OPEX. This is not true. Interest should not go into opex. Now, the reason being is that interest actually isn't an operating expense. This is leverage to be able to accelerate the growth of the business. And therefore, if a buyer or a third party wants to

look at the true profit generation of this business, they do not want interest included in the IBIDA number. Hence, you get interest before earnings, tax, depreciation, and amotization. Then, your interest expenses go down here. And so let's say that there was $10 in interest expenses associated with loans. Then you would have one last line item down the bottom here which would be $10 in net profit. So there is technically two definitions. There's IBIDA which is the capability of the business to generate earnings if it didn't have loans outstanding which is what a buyer wants to

understand because a buyer will just zero out the loans. And then net profit is the actual profit that the business ends up generating after these repayments are made. And this is the basic P&L structure. At a high level, we can also, and this is a very valuable exercise that I'd always do, allocate percentages to each level of the P&L to understand relative percentages throughout the year because we want to be looking at these on a month-on-month basis to be able to understand how these percentages are changing. So, for example, our cost of delivery is 40%,

our gross margin is 60, our me is 20, our contribution margin is 40%, this is 20%, IBIDA is 20%, and then net profit is 10. And so then you can look at these percentages over time and go, are they expanding? Are they contracting? What is our target percentage allocations here? And this becomes how you can use the P&L effectively on a day-to-day or month-to-month basis to be able to make decisions. Okay. If me is creeping up, well, we know our marketing efficiency is declining and that's deteriorating and compressing the bottom of the P&L. So, we

need to go and fix that. If our gross margin is compressing, we need to look into why that's the case. Is this a discounting issue, a returns issue? like what above the P&L is causing the gross margin compression. So the way to look at this is that if a target percentage or a target number on this P&L is not hitting target, the fault is everything above it. So if we're not hitting our IBIT target, it's an issue with either or all operating expenses, me cost of delivery. If we're not hitting our contribution margin target, it's

an issue with me and cost of delivery. If we're not hitting our gross margin target, it's an issue with cost of delivery or revenue. Because remember, revenue isn't just net revenue, but it's also gross. And so there's a few different levers in here that we need to look at as well. So this is how you troubleshoot the P&L. If you're not hitting numbers that you want, you want to look upwards and look at the levers that exist there that we need to pull on and that we need to diagnose. So that is the P&L explained.

Moving into unit economics. So unit economics, how unit economics differs from the P&L is the P&L is a zoomed out view of the entire business. Unit economics is diving into a specific unit. Now what this would typically look like in retail is we would be looking at a single unit. And so if we're selling a t-shirt, we would be looking at this t-shirt and breaking down what is the price, what is the cost of goods sold, what is the cost to acquire a customer for this t-shirt, and therefore what is the net profit on this

individual unit. Now in e-commerce, in direct to consumer, this actually changes a little bit. And we're actually looking at baskets. And so when we talk about unit economics, we're actually typically talking about the basket because an average cart doesn't just have one unit in it. Most people will buy 1.5 things or 1.7 things. That's Called your units per transaction. And so because of that, we want to encapsulate multiple different products into any unit economic calculation. So the single most important component of unit economics is something that we haven't spoken about yet, which is CAC. This

is your cost to acquire a customer. It's one of the most important metrics on acquisition and it's a metric that most brands calculate completely wrong. How you calculate this is you take total advertising spend over any given time period. So this could be on a daily level, a weekly level, a monthly level. And then we divide by the total amount of new customers that we acquired in that period. This is not the same as CPA in the platform because CPA is running off attributed numbers and will end up showing you a better number than is

actually reflective within the business. And then people will also try to delineate CAC down to a channel by channel level, which is also impossible because you're relying on attribution. and attribution has a plethora of issues which is why we have this whole finance video together because attribution is just unreliable. Now the first question that I typically get when I pull a trailing CAC calculation for most businesses is okay well what is a good CAC? You have access to over 68 N figure brands. I've personally consulted on well over a thousand brands to date. I have

a pretty good reference what a good CAC looks like. The answer to what a good CAC is is that it's fundamentally actually the wrong question because CAC means nothing without contextualizing it to a pairing metric. And that pairing metric is gross profit on first purchase because you could have a $100 cost to acquire a customer. But if you're making $10,000 on the order, this is an unbelievable Deal. You're paying $100 and you're making $10,000 on the order. That's insane. And so this could be like super high ticket furniture as an example. Or vice versa, you

could be super low ticket selling lollies online, which is probably not a very good niche, and you're making $20 in GP. This is a terrible exchange. You're actually losing $80 on each new customer acquired here. And so, you need to understand what is your gross profit on first purchase, and it's really important on first purchase. Cuz what people will do here is they'll take their average order value, which on Shopify, let's say it's $100, and then they'll take their gross profit percentage, which is typically wrong, calculated wrong. We spent a lot of time going through

that and they'll go, "Okay, our gross profit on average is 70%. Therefore, we have a $70 GP, so we're happy with $35 CAC." Well, is this on new customers? Because returning customers always have a higher average order value, which pulls your average order value number up. And so, if you actually delineate down into new customers, you might find that this is actually 90 and that your GP is actually even lower because maybe you have more discount orientated front-end offers. And so your GP compresses all the way down to maybe 50. And so now your CAC

targets are completely wrong in conjunction with your true GP. And so this is really just the important of data integrity within the business. These numbers need to be calculated correctly. You need to trust the calculations or else you're just going to scale off fundamentally flawed metrics. So as a part of unit economics, let's go into pricing strategy because pricing becomes an enormous lever for profitability of the business Particularly when products are underpriced. So, as a new e-commerce brand, what people will typically do is they will use keystone pricing. Now, I think anyone that's ever run

a business or started an ecom brand has done this, which is that you take your cost of product, so you go to the manufacturer, product cost $10, and you just double it. This is what's called keystone pricing. So, as an example, a $6 cost of goods just gets 2xed and becomes a $12 price. So many people price this way. I reckon 40% of the startup market just prices in this way. This is probably the worst way you could ever price. Now, it's because this cost of goods doesn't encapsulate all the variable expenses associated with fulfillment.

And so, you end up with way more compressed margins than just 50%. I think the reason why people use Keystone pricing is number one, it's simplicity, but number two, the fact that people don't want to overpric. People think that margin is bad. We can't have too much margin or else we're just ripping people off. But because of that, they don't take into consideration all of the expenses associated to actually fulfill within a business. And so because of that, people underpric and they can never actually scale. Whereas the actual cost to grow a business in all

of the variable expenses associated with fulfillment as well as all the marketing expenses associated and operating expenses taking up 15%. It ends up being a lot. And so you end up needing to price way higher than just a doubling using keystone pricing. So the next pricing model that people jump to is IMU pricing, which is initial markup pricing. So this is effectively the same thing as Keystone, but rather than doing a 2x, maybe we do a 3x or we do a 4x or a 5x. So we're effectively taking the cost of goods once again, and

we're just applying a multiple to this number. Now, that might seem better, right? You're like, well, Nathan said that the big issue with Keystone is you're not increasing by enough. So if we just Increase to 345x, then surely that fixes the issue. It doesn't fix the issue because this is just an arbitrary multiple on a product cost, which has nothing to do with the price elasticity of demand within the market. It has nothing to do with the competitors and it has nothing to do with the actual cost structure associated with delivering that particular product. And

so yes, we could just 4x it now and the price is $24. Amazing. But this might be substantially overpriced compared to competitors. This might still not encapsulate enough margin to be able to actually fulfill on this business model. And so all of these things have to be taken into consideration. The real core takeaways here is that you want to use waterfall pricing from bottom to top to be able to identify the price that you need to actually price the product to make money. And then from there, you make the decision of do we even sell

this product? So here's all the expenses associated with actually delivering on the product. We need to start at the bottom. So what is our actual contribution profit target here? Okay, is it 20%, 30%, 40% based on our existing P&L understanding and how much operating expenses we have. So, you could really start all the way down the bottom here, right? And take this a step further and say that our net profit target is 10%. We know that our operating expenses 10%. Therefore, our contribution target is going to be 20%. Okay, great. Now, we need to start

moving up. How much do we think it's going to cost to acquire a customer on this product? Now, based on our understanding, based on the ad account, based on other products, we think our CAC is going to sit at $30. Now, what's the return allowance? This is also called a shrink allowance, which is that we're allowing for shrink in discounts, Returns, etc. Well, we probably want a 3% return allowance, probably a 3% discount allowance. Shipping and fulfillment on this product, we've gone, we've reached out to Opost or the shipping courier, and we know that this

is going to be $11. Then we have the cost of goods, which we know is $6 over here. And now we can work our way all the way up to pricing. So, we need to be priced at 30 + 11 + 6 plus 3% 3% of this total price. And then we need to come out to a 20% contribution margin. And so we can run all of this math, which I'm not going to do for the sake of this video, but let's say this takes us up to a $100 price because 100 - 6 -

11 takes us down to about 85. Then we minus another 6% which takes us down to 79. Take us off another 30, that takes us to 49. And so down the bottom here, we have a $49 contribution profit, which is actually a really high contribution margin, way above our actual target. So if we just keep reworking these numbers, I believe this should come out to probably like a $70 price. So to hit this contribution margin target with this CAC with these unit economics, we come out to a $70 price at the top. This is how

you price correctly. And you then go and you do competitor research. And guess what? If this $70 price is ridiculous, if everyone else is pricing at $30, $40, well, either number one, we need to figure out, can we position into a blue ocean where no one's actually selling a premium version of this product? Can we position this as premium? Can we sell at this price? And if we can't, we don't sell the product. The unit economics don't work end to end. We can't hit our profitability targets by selling this product any less than $70. No

one will buy it at $70. Let's cut the product entirely and not sell it. So that is how you need to be thinking through pricing strategy. Now you need to understand the unit economics of discounting. This is where Most brands destroy their margin without even realizing it. Now I'll give you a really quick worked example. So if we have a $100 average order value or product price, this means we're going to come out to a $40 gross margin. Now what we can do here to calculate our break even return on ad spend. So what return

on ad spend do we need to be to break even is we take our gross margin which is 40%. So break even return on ad spend equals 1 / 40%. And this equals 2.5. So we need to achieve at least a 2.5 return on ad spend to break even and therefore make money. So this is the lowest we can be. Great. This is a really important number for not only your internal team and your agency to know because if let's say an ad set or a campaign is below this you're losing money. It's really important

to understand what this number actually sits at, but more important to understand where this number sits at during a discount period. So, if we take this exact same example, you're on 30% discount. So, you get slashed to a $70 price. $70 price, your cost of goods stays the same. And so, we have our $60 cost of goods here. That then comes out to a $10 gross margin. If we then recalculate our break even return here, our break even return becomes 1 / 10%, which is 10. So, our break even return on ad spend, the efficiency

that we need to operate is 5x what it was up here off just a 30% off discount, which is crazy. And it's because discounting exponentially increases the break even return or the efficiency that needs to be hit within paid media. And so, anytime you're discounting, it is really important to run this exact calculation to be able to understand what efficiency level we need to be at. Now, all of this Math can also be done based on a CAC or a CPA number. So rather than doing this division to calculate break even return, all you do

is this gross margin number is your break even CAC. So your break even CAC just simply equals gross margin. So here break even CAC is 40. Here break even CAC is 10. Obviously a $10 CAC is absolutely insane. You're probably never going to hit that across paid platforms. Therefore, this discount will never work for acquiring new customers. So this shouldn't go out publicly. Now this isn't to say that you can't discount. You can't do blank discounts. Of course you can. You just need to understand how it affects the efficiency targets within paid and whether you

can actually hit those efficiency targets. You always need to just run that math which is that if we're going to do a 20% off discount, how does our efficiency on paid media need to change to maintain the same level of profitability or improve profitability? And if it's insane, if it's like we need to go from a 2x row to a 12, well that is physically impossible. The discount isn't going to create that much of an uplift in demand and conversion rates. Therefore, we need to rethink the discount approach and what we're actually doing with this

offer. And so, let me give you some other options. Well, bundling is a very big one. Everyone knows this. This is pretty generic, which is that if you bundle, you get economies of scale because you might go from one unit in the cart, too. But this doesn't increase your shipping and fulfillment cost from 1 to two. Generally, you will get maybe a 20% inflation in the shipping and fulfillment cost of the product. And so, therefore, you actually get better gross margin when you bundle. And you can give that margin away to the customer, and it

doesn't affect your GM percentage. Bundling also increases average order value substantially, which allows for you to have a higher cost of acquiring a customer on the platform and still make More money. And so bundling is still a really effective way to provide a discount to the end consumer, provide an offer that seems to have a value discrepancy in the market that allows someone to get over the line and buy, but it is beneficial to you in regards to the unit economics of that bundle. Another example here is a gift with purchase. And so this is

where you just need to become good at calculating unit economics correctly on offers because this is what allows you to validate whether an offer will work. It allows you to bake in the assumptions and then you can go and actually run it in public. So an example of this would be if you buy two beach towels, you get a free bag that actually holds the beach towel. Now the beach towels might be $100 each is what they're priced at and they're pretty high margin. Let's say 70% or something. So you go and when people buy

two, it's $200, but they get a free bag. And you can say the free bag is valued at or sold on the website at maybe $50, $60. You can price it really high. It can be seen as a premium bag. Uh but the reality is is that the cost of goods on this bag for you is maybe $4. And so you're going to forgo $4 of margin to get an extra $100 in revenue. Really good exchange. Okay, your gross margin is pretty much going to stay the exact same as a percentage, but you're doubling average

order value by giving this free bag away. So gift with purchase ends up working really well if the offer is crafted well and if the gift is relatively low cost. And then the last is straight uh discounting which I just told you compresses margin a lot and you have to be really careful about but the caveat here is that you can do this on grade C inventory. We'll talk about stages of inventory later in this video and we'll break down actual strategies to be able to move inventory And focus on cash versus net profit and

the marketing strategies associated. But just as a call out, if you are going to do flat discounting, you want to do it on inventory that isn't moving. And then you can move that inventory back into cash, even if it is at break even. It doesn't really matter because the inventory wasn't going to sell anyway. Now, I can't talk about unit economics and finance without talking about LTV and repeat purchase rates. And the reason this is so critical is because acquisition is expensive. Fundamentally, it is expensive to acquire customers. Particularly if we start going into an

industry like CPG, consumer package goods. you're not going to probably even be profitable on first purchase. And the reason being is that you just have competitors that will outspend you, outbid you at auction, on Meta, on Google, and they can go and pay $200 to acquire a customer because they have this massive lifetime value that they can then realize on second, third purchase. And if you don't have that, you will lose because your competitors will just spend more than. And so lifetime value becomes a critical component in profitability of most business models. I can give

you a completely different example of this outside of e-commerce as a whole, which is actually the agency model, which is something that I'm very familiar with. In the agency model, most agencies won't run profitable on first purchase or first invoice on month one. Most agencies will run at a 3 to six month CAC payback period. So, they will spend X amount to be able to get a client, whether that's associated with marketing, sales costs, etc., etc., networking, events, and then once they have a client, they will only start making money after 3 to 6 months

due to the CAC. Same thing in CPG, same thing in fashion, retail, etc. Except in these Other industries, you need to carefully and meticulously understand what your LTV actually is so that you can operate that model profitably. Because if you are an agency, let's say, and you don't understand all the associated cost to get the customer and you miscalculate CAC or you miscalculate LTV, you are actually in a very different profitability position and you can really negatively impact yourself. So, let's run through a quick example. Let's say you're doing 200k a month in revenue. Of

that 200k a month, $140,000 is new customer revenue. And let's say this is coming from 50k in ad spend. That then means you have $60,000 in RC revenue. And let's say that the cost here is like 2.5K, which is associated Clavio costs and maybe an agency associated with driving repeat revenue. Now, your new customer profit contribution, which isn't a metric that we've gone through yet, but it's just contribution margin, but we're doing it on new customer revenue only. This is 140 minus 50. So, we have $90,000, but we obviously also need to minus cost of

delivery and gross margin. So, let's assume a 50% gross margin. That means that this is going to come to 70 - 50. This is going to be $20,000 in profit contribution that we're making each month on new customers. Now, what about RC revenue? Well, 30 - 27, this is 27,500 on return. And so this business, which is a pretty typical business in terms of new customer revenue to returning customer revenue percentages at this kind of size, this business is making More on returning customers in profit per month than they are on you. And this is

super typical because majority of your profit ends up coming from returning customers. Now, why is that the case? Well, because it's very hard to get a new customer. You have to pay money to get them. Returning customers come back due to the product, the experience, and the brand affiliation. So you actually don't need to spend much at all on getting a returning customer and they drive tons of profit contribution in the business. Now once you start getting into this position, that's where you start needing to understand, well, can we actually push up new customers even

harder and subsidize this acquisition cost with all the profit that we're making on returning and continue to grow the business that way. And that's where I just want to provide a massive hesitation to most people, which is that most people calculate LTV wrong. Number one, they will calculate LTV based on just infinity. And so when we look at lifetime value, uh, as you extend the time period in which you're looking at how much a customer is worth to you, it increases forever. And so we can look at how much a customer is worth to us

after 6 months and it might be $100. And then we can look at 12 months and it might be $10. Then we can look at 18 months, it might be $120. And as you just keep extending that time horizon, LTV increases forever. Now it does somewhat asotope. So it will look something like this. But still, if we measure from here to here, there is still an increase. And so what you want to be very careful of is not just looking at an LTV calculation based on total customer data over forever and instead you want to

restrict it to a particular time period. Now a good way to do this is to look at 90day or 180day LTV. The second delineation you want to make is you want To go to LTGP. You want to be looking at gross profit, not value. Now, LTV in the traditional value sense actually is lifetime gross profit. When we measure this against software because this is a software metric and in software, lifetime value is typically very close to lifetime gross profit and therefore it's the same thing. But in e-commerce, we want to make this clear delineation because

some people and a lot of softwares actually will give you LTV as a revenue number. And so we want to understand what the gross profit are of these customers at 90 180 days. And then we can use that to be able to understand what we can actually pay to acquire a customer. Now, let's take unit economics and apply it at a product level. So, on product level economics, let's say you have a hoodie that you sell and you sell it at $90 retail with a $21. So, you have a hoodie at $90, cost of goods

21. So, you have gross margin of $69. Our cost to acquire a customer through the hoodie on our advantage plus scaling campaign is $45 right now. So, that's our CPA. We're just going to call it CAC. So, our contribution margin here is $24. Now, we have a t-shirt that's priced way less. Cost of goods is about the same. Actually, our gross margin is only 27, but we have a CAC of 15 and we have a contribution margin of $12. So, right now, the hoodie is driving 2x the contribution margin to the business per sale. So,

this firstly is an important number for us to be across. Okay, understanding what actual products within the portfolio is driving us the most contribution margin when we break our unit economics down at a product level. The second component that becomes important here is understanding how we should structure our campaigns accordingly. Should we be splitting These out? Because if we're using a bidding strategy on Meta, Google, Tik Tok, which is maximize conversions, and this is the default bidding strategy on every advertising platform. When you use maximize conversions, what it's optimizing for is the lowest CPA. So,

whatever product is driving the lowest CPA, whatever ad is driving the lowest CPA, that will get prioritized, that will get the spend. Now, if we look at this example here, the t-shirt actually has the lower CPA. And so, Tik Tok, Meta, Google will distribute all your spend here, but this product is driving a lower contribution margin per sale. We would actually prefer Meta to put all of our spend up here. Yes, it's going to cost us a little bit more money, but we're going to drive more contribution margin per product sold. So this is the

better place to put cash right now. Maximize conversions won't do that. Now you can fix this in the platforms in a number of ways. You can have segmentation across product categories based on product level economics which is really important to do. You can change the bidding strategy to maximize for conversion value and therefore we will actually prioritize this because it has a better rorowaz. This actually has a worse rorowaz too. So maximize conversion value won't fix this issue either. So segmentation is the way that you actually need to fix this. Now the other reason why

this is an important exercise is particularly in fashion what you were trying to do is sell through all of your product highest contribution margin possible. What we need to start thinking about in this instance is this t-shirt will this sell anyway without us even pushing it and paid cuz the cost to acquire is so low here that it seems like there is some kind of virality component. there is some kind of product market fit that's just getting this T-shirt to sell kind of regardless of our paid media spend because this is so hyperefficient. So the

question we need to ask particularly if we're an omni channel business with retail stores or we're just a very large business in general is would people have bought this product anyway if we didn't spend? And if that's the case well then guess what let's not spend. Let's then take our contribution margin up on this order by $15 to $27. and let's actually reallocate all the spend to the hoodie which might not be selling naturally or organically and we need the paid spend to be able to drive sellrough rate. So this is where product level economics

becomes incredibly important. It's something you should be breaking down. You can obviously build Google sheets around this. Have visibility build it into your decision-m. Now moving into the metrics that matter. How we think through this internally is through a pyramid. We have platform level metrics down the bottom. We then move into finance grade metrics which we've been covering a lot in this video. And then up the top we move into incrementality testing which probably isn't applicable for 80% of people watching this but for the 20% that's doing over $10 million a month that has a

presence in the US or multiple different markets might be omni channel this is going to be critical for ensuring measurement within the business and it's going to tie into understanding how finance ties into actual platform metrics. So if we start down the bottom we have rorowaz CPA CTR multi-touch attribution like triple whale etc. Now why is rorowaz down the bottom of the pyramid? Why is rorowaz not good? The reason why return on ad spend isn't a good metric is not because calculating ROI is bad, calculating return on investment is a great thing to do. The

Reason why rorowaz is unreliable is because it relies on attribution. And attribution is fundamentally an unknowable reality where you're trying to connect correlation within platforms with faulty data to be able to prove causation. And what do I mean by all that complex language? When someone gets served an ad and then they click on the ad and then they go to the website and they purchase. You might think that this is causation that this ad click caused the purchase. But in a lot of cases particularly in large omni channel retailers that actually is not the case.

This is correlation and in fact all attribution is just correlation. We are saying that these two events are correlated. The traditional way to explain correlation versus causation is that at the same time throughout the year, swimming deaths go up and ice cream sales go up. And you go, well, is swimming deaths causing ice cream sales? Is ice cream sales causing swimming deaths? No, they're not causal. They're correlated to summer. When it is summer, more people die when they're swimming in the ocean and more people buy ice cream. And it's the same thing here. Ad clicks

is not necessarily causal to purchasing. It is simply correlated. And so because of this return on ad spend ends up lacking validity and congruency to the P&L. What you see a lot of the time is that people will go and say rorowaz is unbelievable. Wow but my business is dying. What is going on? And it is because rorowaz has a few dynamics to it. Number one because of the way that it attributes. It will always overattribute to the bottom of funnel. Because people that for example might see a billboard up here and this is obviously

a super extreme example but let's say someone sees a billboard and they become aware of uh the brand. They Then see a TV ad which reinforces that they actually really want to buy your stuff because there was an influencer in there who they connect with. Then eventually they get an ad on Facebook. They click on the ad and they buy. Now did that ad on Facebook cause them to buy? Probably not. Like yes it got it over the edge at that specific moment but they were going to buy regardless at the next point of activation.

It was the billboard on the TV that actually warmed them up at the top of funnel but these got no attribution. Now, it's the same thing across platforms. And so, a really evident example of this actually right now for us, and this might change depending on when you're watching this video is Tik Tok to Meta to Google. Now, this isn't applicable for everyone. This is a very unique circumstance, but it at least gets the point across, which is that we're finding at the moment for some very large retail brands that Tik Tok is actually underattributing

substantially because it's getting impressions. It's getting in front of people in new markets, but they don't actually click off the platform much. or if they do click off the platform, they click off, they look at the product page, but they don't buy. Then once we have them in the pixel, Meta goes and follows up and retargets them. And we end up capturing a lot of increased demand on Meta when we increase our Tik Tok spend. And then finally, people that still haven't bought after Tik Tok and Meta, they end up going to Google. They search

for the brand name, they click and buy. And so when we increase Tik Tok spend, our meta rorowaz goes up and our Google rorowaz goes up. Tik Tok doesn't. If we're using rorowaz as the indicator for performance and budget allocation across the business particularly and this is where this becomes a very large issue is when this is reporting up to a seuite a seuite will see return on ad spend numbers across these different channels and go okay tik tok we should cut it's not driving returns meta is okay let's decrease budget and team let's put

more budget into Google that would be a fundamentally terrible idea Because what we're effectively saying there is let's cut all the top of funnel generation and move to just bottom of funnel bottom of funnel will stop working unless there is top of funnel generation. And so rorowaz ends up misleading people in terms of decision-m unless there's a lot of additional nuance in understanding what is actually driving impact in the business. And so we need to go to better measurement systems to be able to actually encapsulate this, understand it, and then communicate it to a seauite

or if you're a small business just to you as the founder. Now we could go on and on about the other limitations of attribution and return on ad spend here, but this is fundamentally the crux of why this metric misleads you in terms of decision-making. So then we move one stage up in the pyramid and we get to financial grade KPIs. Now these are metrics like acquisition me. You can throw me in here as well. We've got profit contribution which we've ran through. We have CAC which we've ran through. So these are all metrics that

aren't relying on the inplatform attributed numbers to be able to calculate them, but instead they're using the actual financial metrics that exist within the business. So the actual amount of new customers that you're acquiring, the actual new customer revenue, and then we're dividing by the actual ad spend. So this is all metrics that aren't relying on any kind of correlation calculation but instead are directly associated to financial metric. And so because of that the reliability is much better. As we move up the pyramid reliability improves however speed decreases. And so what ends up being the

case with these metrics is them going up and down. You need to look at this on slightly longer time periods to be able to make correct decisions whereas you will get faster feedback loops generally speaking on your inplatform metrics. So why would we even use this bottom half of the pyramid at All? Why would we even look at rorowaz CPA? They are helpful in directional campaign feedback and ad feedback within the siloed platform. So what we do not want to do, what would be a big mistake is comparing a rorowaz number on a Google campaign

to a rorowaz number on a meta campaign because they have different attribution models. They're sitting at different points in the funnel. They're driving different incremental impact to the business. If we took both of the campaigns and we pushed budget up, one of them would drive excess return compared to the other irrespective of the rorowaz number on the campaign. So these are helpful directionally within the own platform. So if we're looking at two meta campaigns next to each other, we can compare rorowaz, we can use that to make decisions. But if we start comparing platforms, if

we start zooming out and making larger business level decisions based on these numbers, that's where we can start to mislead ourselves and go into using low reliability, low data integrity level metric. And then all the way up the top here, we have incrementality testing. We won't dive into this in too much detail in this video, but the crux of the way that this works is that you take an area, generally the whole country. You split down by states. In Australia, you have to split down by commuting zones rather than states because there's not enough state

selection. And from there, you hold out a particular area. Now, typically it might even be two states. So, we can go and grab two different states that are next to each other. And then we hold them out. Now, a hold out means that we increase spend everywhere else except here. and we see what is the differential or we can just increase spend here in the control group or we can cut spend entirely. So we have a lot of different options in terms of test design here but the idea is that when we isolate a large

control verse treatment group we can start to measure the incrementality of changing different Campaign types. And so we might go and take that Tik Tok campaign I was talking about before that doesn't seem to be doing well and we might double budgets in these states. And then over a 30-day period, we measure the revenue realization difference over that 30 days and maybe new customer revenue doubles. And so as a function of that, we can take the lift in new customer revenue against the control. We can do an incremental return on ad spend calculation and we

end up getting a very accurate read on the impact that these campaigns are making. Now that is an unbelievably simplified explanation of how this test design actually works. It's actually a lot more complicated. I could make a 2-hour video just on the data science and approach to building incrementality tests, which I think we actually might do at some point. That is the crux of the idea as to how it works. I do not recommend that you go away from this video and think, "Oh, that was a really interesting idea. Let me just make a Google

sheet and start doubling budgets in states." If you do that, you won't get good results. It'll pretty much always say states aren't incremental because the test design isn't set up correctly. You'll end up way overspending or under spending. You won't get statistically relevant results. you need at least a fundamental understanding of data science to be able to start to approach this portion of the pyramid. Um, but that is ultimately the most reliable option. It's just very slow. You can't run a lot of them and you're quite restricted. Now, I said before that I was a

little bit hesitant, put me in here. And the reason why I was a little bit hesitant is that me is revenue divided by ad spend. Sometimes people will do this the other way round and they'll do ad spend divided by revenue, which will give you a percentage number. This instead will give you a multiple number. And so if we for example have $100 in revenue and $50 in ad spend, our me would be a two. So every dollar we spend on ad spend, we get in revenue. You could also look at this as an ROI

calculation. Now this looks really good and this looks like a Good way to index the performance of total spend across all of our media channels against total revenue. But the reason why it's fundamentally a terrible metric for indexing the performance of paid media over time is that it includes a large bucket of revenue that has very little to do with paid ad, which is returning customer revenue. And so we really want to strip returning customer revenue out of this calculation because we don't want marketing taking credit for all these returning customers that would have returned

anyway. So that's why we always want to delineate all of our finance grade metrics into new customerbased metric. And so rather than looking at me, we want to look at a me which is new customer revenue divided by ad spend. And this might actually give us in this example a 1.4 which might not actually be profitable at all for us. And so we might want to be rethinking our whole acquisition strategy based on this delineation to a better number. Now there is a metric that's better than everything that I just put down in the finance

grade section of that pyramid. And it's because I wanted to spend some time on it by itself and give you benchmarks. Now this is LTGP to CAC. Similar to what we talked about before when we had CAC and I said that we needed a pairing metric to be able to contextualize it. This is contextualizing those two numbers. So we're taking the gross profit on a customer and we're dividing by the cost to acquire that customer. And this will give us an integer. And so as an example, if we make $100 on a customer, it costs

us $33 to actually get that customer. This would be 100 divided by 33, which is 3.33 LTGP to CAC. Now, where we need to be really careful with this metric is that once again, just having an unrestricted time on lifetime value is a terrible idea cuz this can be measured across 6 years and so we end up way overspending, but we don't actually realize this cash for like four years into the future. And so you want to put a time constraint here. Now generally what we will do is we will measure this Across two time

horizons. We'll measure it across 30-day LTGP to CAC and we will do 90day LTGP to CAC and then sometimes we'll do 365 days as well depending on how aggressive the acquisition strategy is. Now as benchmarks here what we want to be seeing on either of these metrics depending on how aggressive the acquisition is in the business is that there is multiple different levels that you can be at. So you can be at sub one, you can be at 1 to two, you can be at 2 to three, you can be at 3 plus. If you're

under a one, this is generally speaking a terrible position to be in because it means that you are actually losing money on acquisition. You are losing profit because you're paying more to acquire a customer than the gross profit on first purchase that you're making or within the first 90 days. You have to have incredible retention after this point to be able to support losses on acquisition. Most brands that I audit that are losing on acquisition do not have good enough retention to support it. So this typically is not an actual acquisition strategy most of the

time. This is poor efficiency on acquisition. So this is simply a fact that your ads in your acquisition funnel isn't good enough and this is a bad position to be in. Some people 0.01% of people can get away with this and they have the finance capability and the modeling to be able to actually uh operate at this scale and they have really good LTV. Most people can't do this. You do not want to be below a one on 30-day or 90day. It is worth noting that being below a one isn't necessarily just a CAC issue.

It isn't necessarily just poor efficiency on acquiring, but it could be poor uh gross profit. So, if you only have like, let's say, less than $70 in gross profit on first order, that's probably a problem. You just don't have enough Gross profit to be able to substantiate acquisition at scale on a paid platform. We then have one to two. This also isn't a great position to be in for most people. Now, if you're in CPG and you have really good retention dynamics, and I'll give you some benchmarks on that later on, then you can operate

in this area. But for most people with mediocre retention, that's okay. And if you think your retention is good, it's usually mediocre. You will know if you have excellent retention because the numbers become very evident and you can push acquisition like crazy. So one to a two. You also generally don't want to be here. This is generally not good. Dash average. If you're really wanting aggressive growth, if you're financing hard, if you have investors and you need to just uh throttle up revenue, sure you can scale on this, but it's not ideal. 2 to three,

this is optimal. This is where you want to be. This is a great zone. you should be scaling up budgets. Now, to give you a little bit of context here, what does a two elig look like? Well, if you have $100 in gross profit on first purchase, that's simply a CAC of 50. And so, take whatever your gross profit number is on first order, divide by two, and that's what your CAC would need to be to hit this optimal range, to hit the bottom end of the optimal range. Now, greater than a three, this is

actually also a big mistake. Being less than a one is is a mistake. Being greater than a three is a mistake because this means you're just leaving money on the table. you could scale very aggressively here, be unbelievably profitable, and you're probably underleveraging paid media or whatever marketing channels you're using to drive this efficiency. Now, let's add two quick notes here before we move on. Note number one regarding attribution and incrementality and everything that we Discussed on the pyramid. Uh when it comes to attribution, what you want to make sure of is that you're not

using view through conversions in the platforms. And so when you're looking at meta specifically, you want to be making sure that you're using 7-day click as your optimization or your reporting. This is going to give you much tighter congruency to acquisition me. In fact, in most businesses, when you look at their acquisition me and then you look at their 7-day click rorowaz on meta, it is super correlated and that's because it doesn't include all these viewrough conversions that is overattributing in the platform. For those that don't know what a view through conversion is, it's when

a user sees your ad on Facebook, doesn't click, but then buys within 24 hours, Meta can claim the conversion. And a lot of those people are going to buy anyway. They're existing customers. List goes on. And so, you want to be not including them in your reporting. You also don't want to be including existing customers. As I said before, existing customers aren't that incremental on the platform, and so you don't want to be overspending here. You want to go to breakdown audience segments and look at where your spend's going. Often, people are putting way too

much spend to existing customers, and their frequency is way too high. So you want to pull out the frequency column and you want to make sure that over the last 7 days it's under an eight. Any higher than an eight and you're definitely overspending because you're serving to existing customers more than eight times a month which is not incremental. So there's a big existing customer trap on the platforms. The platforms always want to spend here because they know they can overattribute and they always want to expand their attribution windows because they know they can overattribute

and if they can attribute more revenue you will spend more. Now, as a bit of a formula here, cuz I get this question a lot, even from 700, $800 million brands. I get this question when when we come in And talk to them and have calls with them, which is what percentage should we be allocating to existing customers? You're saying that existing customers aren't that incremental. Well, so then what percent should we be allocating of our budget? Should it be 20%, 30%, 40%? Well, because 80% of our revenue as a large 9 figure brand is

coming from existing customers cuz we have almost full market saturation in Australia. So, what are we doing? The question isn't percentage. Thinking about percentage allocation of ad spend to existing customers is just a bad way to look at it. What we instead want to look at is total amount of existing customers. So how many existing customers do we have? How many times do we want to show them an ad? And then from there we can calculate how much we need to spend because as long as we know the CPM we can get the total spend

per month. So I'll give you the formula and I'll give you an example which is let's say you have 300,000 existing customers and you want to serve how many ads to them? Let's say you want to serve three times a month. Then you just times by your CPM. So go into the platform, do an audience segment breakdown, look at what's your CPM on existing customers, and let's say it is $2.50. Now, that's super low. It's probably going to be a lot higher than that, but let's just say that for the sake of this example. That

means that you're going to have to spend $2,250 per month on existing customers to hit them three times. Now, you can go and change these variables and that will change the outcome and tell you how much to spend. Reality is you need to spend way less than you actually think. Most people think, "Oh, we're a massive brand. We have multiple millions, if not tens of millions of customers. We need to spend hundreds of thousands a month targeting them." You typically don't. You could hit them with a 3 to six frequency and you could just spend

10, maybe on the upper side, $30, $40,000 a month, and you're completely fine. I see tiny businesses spending $40,000 a month on existing customers. And so, this is the math. This is what you want to be doing. Let's move on to the final topic here on metrics that matter, which is the profit frontier. as a founder or a head of digital or a head of marketing, the question you need to be able to answer is, if we were to spend an extra $10,000 next month, where would we put that budget? Where would we get the

best incremental impact of that media spend? Most people don't know that answer reliably enough. They don't know actually where they should be allocating their spend and therefore they get into a position where they continue to drive spend up across platforms, but they don't see incremental returns. And so how you need to be thinking through this problem is what's called the next best dollar. And so we want to have all of our platforms, let's say Meta, Google, Tik Tok, add in a bunch of other channels if you're spending on them. Ideally, you shouldn't be spending on

a bunch of tertiary channels, but let's say you are. And then if we just take spend here, we want to take Meta up to the level of spend in which we stop getting returns that we couldn't get on the other platform. So let's say this is the current allocation of spend. Okay? Maybe we're like 100,000 here, 50,000 here, 70,000 here. What we want to know is if we went and put an extra 10,000 into Tik Tok off the top here, what revenue return would we get of this $10,000? Same thing for Google, same thing for

Meta. And this is what we're constantly trying to solve for because ultimately to grow the business, we need To continue allocating more marketing dollars across the current marketing channels. And we need to do it in the most efficient way possible. The only way to really understand this is to unfortunately run incrementality tests. We would want to for example go and put 10 into meta but do it in a controlled incrementality test where we can get a read on what this is. Same thing for Google, same thing for Tik Tok. Now we can also just do

this intuitively over time. You could use MM as well, some marketing mix models to be able to identify where you should be putting media spend at a lower revenue threshold like seven figure brands. This should be relatively intuitive if you're a good performance marketer. But this is ultimately the problem that you should be solving for. And you should also be thinking about this inverse. And so, could we pull 10,000 out of Google, reallocate it to Meta, and get a better ROI? Maybe this 10,000 here is only driving us a 2x, but we could go and

put another 10,000 into Meta, and it would give us a 4x on new customer revenue. And so, we should be rebalancing budgets and reallocating and scaling across here based on incremental returns and the profit frontier. And so, the profit frontier is that you want to go all the way up to the point in which you're making no more profit past this nominal dollar in spend. And so you want to find the point in which your acquisition me becomes the break even point. And once you hit that break even point incrementally, that's where you stop spending.

And you're doing that across all channels at all times. And only until you've reached the profit frontier on all of these three primary channels do you go and move on to adding in additional channels. So cash flow verse profit. Profit doesn't equal cash. And that's ultimately why paid media within e-commerce is way more complex than in a service-based business or than in a SAS business or particularly in the info space. And it's because when you scale Up, you need to commit cash to future inventory purchasing which substantially restricts the actual dividends that could be yielded

within the business. And so if you see an e-commerce business doing 100k in profit, reality is founders probably taking no money. even a million in profit, $2 million in profit, respective to the total revenue and growth rate, there may be actually no profit available at the end of the day cuz it all gets reinvested into future inventory buying. So let's say that the P&L here shows $2 million in profit per year, but the balance sheet has $3 million on it. So the balance sheet will show you the assets of the business. The assets in e-commerce

is typically unsold inventory. Now there could be buildings on here if they own the office, if they own the warehouse, etc. But typically for most businesses, we're just talking about inventory sitting on the balance sheet as well as cash sitting on the balance sheet. So the reality of this business right here is that they are not in a good position. The brand actually has negative cash flow depending on how we're looking this across time and how the cash is actually moving. But if they made $2 million in profit, but $3 million in inventory didn't sell,

well, they actually made no money. They would actually be cash flow negative. they would be down a million dollars in cash because yes, they made 2 mil, but they bought 3 million in inventory and it never moved. Not good. So, this is where we get into the inventory death spiral or what can also be called as skew rationalization. So, all of your inventory has a grade associated to it. And most people are probably aware of this, but the marketers and the performance marketers Watching this aren't. And so, this will become incredibly helpful in making paid

media decisions and tying this into how we should be approaching the ad account. Grade A is fastm moving inventory. This is inventory that is selling quickly that we honestly don't need to worry about. We need to just be thinking about how do we not go out of stock. Grade B inventory is medium sell to rate. This is slowm moving sell to rate and then this is not moving at all. Now Shopify actually auto ranks your inventory anyway as long as the inventory is within Shopify. You can actually see this yourself. I think it's under the

products tab. Now the key here really is that people will do product launches because product launches are one of the best levers, one of the best ways to scale an ecom brand. Ultimately you have brands like Grunes, like AG1, like IM8, all of these brands have gotten enormous, hundreds of millions of dollars, close to a billion dollars with just one product. But that is misleading you in terms of how most e-commerce brands need to grow because those brands have done so well because they're in the supplement space. And in the supplement space, you have a

very unique advantage, which is that you can reposition your product into technically like different products. So, if you're selling a multivitamin, you can say that that multivitamin helps with gut health, but it also helps with hair health, but it also helps with all these other problems. And so, because of that, you have a really large total addressable market that you can reposition the product into. With most brands, you cannot reposition your product enough to gain a large total addressable market. So, you need to do future product launches. Now, when you do that, what ends up

happening is they don't work. A lot of product launches fail. And so They either land into grade B, grade C, or you get an initial pop from organic and existing customers and you can sell through maybe the first 30 40% of inventory and then it doesn't move at all. And so then you get stuck in a situation which is that every time you're trying to grow through product expansion, you were just adding more cash onto the balance sheet that is restricting the cash position of the business and the ability for you to scale. This is

where skew rationalization becomes incredibly important, which is that as new products are introduced into the product suite, you need to be thinking about how can we rationalize our skew count down always to be able to decrease the amount of available products, to be able to decrease the amount of inventory on hand. There also needs to be some congruency here between the marketing team and the finance team. And this is where an enormous mistake is made, not only on agency side, but internally a lot of the time, which is that let's play through an example of

this, which is that you go and you launch a new product. it doesn't do very well. Okay, it works on the organic list. You get a bit of a pop out of it. Maybe you do 100k, but you ordered a million dollars worth in sell value. And then because of that, it ends up over here in grade D or grade C. It's barely moving. You look at the sell through rate and you go, we currently have 400 days of inventory on hand, which means it's going to take 400 days for us to sell through all

this product, which isn't good. Okay, we want to move this back to cash as soon as possible. Now, what will normally happen is the CFO or the owner or the manager or whatever it is, will look at this and go, "Ah, that's annoying. and then try to solution it themselves. But a lot of the time, we can actually just solve this through paid media. Okay, we can take this product that didn't do too well and we can start pushing spend behind it on paid media at break even or maybe even a slight loss, dedicate isolate

it into its own adfunnel. So, it doesn't even have to sit on the website. It can sit separately. We can push it through whitelisting pages. What we've also done for a client is we've actually taken their grade D inventory and we've pushed it into a new country that we knew would perform well. we could discount heavily, wouldn't erode the brand equity within the primary market, and then we can just move all this stock using paid and turn it back into cash. Now, people don't have this conversation enough with the paid media team because the paid

media team, the agency, is always KPI on profit, and this can be really detrimental. If we're trying to just maximize the profit position of the business, what ends up happening is all of our spend will actually go up here and we'll end up really messing the business up because they'll have all this grade BCD inventory that doesn't actually get prioritized in paid media at all cuz it doesn't have a good ROI, a good row, a good efficiency. We need to push grade BCD inventory in paid to be able to turn it back into cash. This

is where you do need an understanding of the inventory position of the business so that we can make better decisions on paid for the overall health of the business. All right, we can't talk about finance without talking about cash conversion cycles. The actual formula for cash conversion cycle is DIIO plus DSO minus DPO. So days in inventory, days sales outstanding, days of payable outstanding. Now this is a 20-minute video on itself. In fact, we've actually put out a lot of content on cash conversion cycles, how to calculate it, how to measure it, what's good, what's

bad, etc. So instead of going too deep here, I instead just want to explain the top level concept so that if you're in marketing, you can understand how this actually relates into the business and where you might need to be understanding of these components. If you actually own the business or you're a CFO, I Recommend watching our other videos that go into a little bit more detail on this topic. But the fundamental idea here is that if you sell 10 units of product and you make $50 in gross profit here, and once again, I'm just

making numbers up, but this will make sense. What ends up happening here is because you only have $50 left in your hand, you can only then go and buy 11 units on your next purchase order. And then you'll make $55 in gross profit. And then this will allow you to maybe buy 12 units on the next order. And then this goes on and on and on. And so the limiter in this business is not the actual selling of the product. It isn't marketing. It isn't ROI on Facebook ads. It's the fact that they're just cash

limited in their ability to buy future inventory because the amount of gross profit that is generated off this purchasing only allows a slight increase in future purchasing. Now, what you need to also add in here is let's say that there is a 90-day lead time from putting the inventory in to actually arriving at your warehouse to when you can start selling. And let's assume that you have very poor terms with the manufacturer cuz these POS are so small. So, you actually have to put the money down 90 days prior to the actual inventory coming. then

you're in a terrible position because you have to buy these 11 units after you sell these units to get the cash. And so you actually then have to go out of stock for 90 days here cuz once you have this cash, then you can buy the units, then you can actually go and sell again. And so what you would want to do here in this example is you would want to negotiate on better supplier terms so you aren't out of stock for 90 days. And so you might actually negotiate that you only need to pay

20% up front here. And so after you sell two units, you have enough money to put in your next order. And then as you're selling through these other units, you're going through this 90-day period. Then once these actually get shipped, you pay the rest. And so maybe the shipping time is 20 days. You're actually only out of stock for 20 days Rather than 90. Now, another thing that you could do is you could reduce the lead time from the manufacturer. So rather than it taking 90 days, you could try get a different manufacturer that can

reduce this down to maybe 30. And then that fixes the whole problem in itself. Or the last one is you use uh some kind of inventory financing. And so you actually take a loan out to pay for this inventory so that you can sell through, get this profit. This profit pays down the loan as you're going. And then this inventory comes. You then start selling this inventory. You open up another loan and this loan pays for the next purchase. And so you can get ahead by opening up some kind of financing. Now the issue is

that financing is very expensive. When companies are loaning, you'll pretty much always have to put yourself up as collateral as well. And it's super risky. anytime you're introducing a lot of debt into the business, you're banking on the fact that you could sell these future units at an efficiency that's going to allow you to pay down the loan. And so, if you are agency side and you're working with like seven figure businesses, they're almost always using some kind of financing to be able to actually grow. Um, and you need to take into consideration that rapid

growth for these types of businesses usually actually isn't possible. The limiter isn't on how quick we can increase ad spend or how efficient the ads are. The limiter is the cash conversion cycle. And so I have actually seen single-handedly probably five times now seven figure businesses that have gone to an agency. The agency has absolutely crushed it. They've ramp spend from $20,000 a month to $200,000 a month. They haven't actually had any visibility into any of this with the client. The client in the background to fuel this growth has gone and taken millions of dollars

out in inventory financing to be able to fuel the growth to be able to buy all the future inventory required to hit that kind of tripling per year in growth rate. And then as they've done that, their Interest repayments get so large and the ads start becoming inefficient to where net margins get squeezed down to zero. And now, yes, we've taken a business from doing $2 million a year to $7 million a year. Like, amazing. Good job, agency. But the business has gone from cash flow positive, founders actually taking an income to the business is

near zero on net profit after all the interest repayments and the compression in marketing efficiency as they've achieved that additional scale. And so this is why this is so important for at least people to be across so that they can understand that you just like can't triple an ecom business in a year. You just can't because the cash requirements on inventory are just so enormous. And so you either need a lot of self-funding. So the person starting the business needs to have multi-millions liquid. They need funding from someone who's providing the multi-millions or they need

to go into some kind of financing. Otherwise, you just can't scale inventory orders fast enough to keep up with the scale that you want to achieve within the business. And a lot of the time where this gets a little bit dangerous is that uh people will pull marketing efficiency down to achieve larger scale but to do that it yes it accelerates the growth rate but it also increases the amount of debt that the brand needs to take on which actually compresses margins down to zero. And so you would actually be better off growing slightly slower

at a slightly lower me. So a better efficiency, less marketing spend, having less interest repayments and making way more money rather than just trying to arbitrarily accelerate your growth rate but increasing interest payments and making marketing efficiency worse. So that's just where you need to be very very careful in your financial modeling. Now this is where ultimately there starts to become a CFO verse CMO disconnect in a lot of businesses in this core decision which is let's run Through an example. February arrives and inventory from November is still unsold. Maybe there was some new products

in here that didn't do so well and they've fallen into grade C and D inventory. Now, most CFOs will actually look at this and go, "Okay, we need to cut media spend because we actually don't have enough cash right now to be able to fuel the current marketing spend in Feb through to April." And so, let's go and take our marketing expenses and cut them by 40%. And now this is quite logical when you're just opening up and looking at the P&L because when you look at the P&L uh cost of delivery might be 30%,

marketing might be 25% and then opex is 15 and so profit in this business is at 30%. Now when revenue suddenly decreases in Feb and we actually don't have a lot of cash available me might go up and spike to 30% which starts compressing profitability down and profitability dips to 25%. Now naturally the reaction from the CFO should be this has gotten out of control. ME has accelerated up. We need to cut marketing spend. Marketing team cut your budgets. Now there's two issues here. Number one is that in this business it might very much so

be the fact that the marketing is driving the revenue. And so this compression in marketing efficiency is an efficiency issue on the current media spend. We shouldn't be cutting media spend because if we cut media spend, revenue will dip further. And so you can get into these uh cyclical situations where marketing spend drops, revenue Drops, profitability decreases. To fix profitability, marketing spend needs to drop even more. And you keep dropping marketing spend to try to fix the issue, but it doesn't fix it. The other issue is that to fix this grade CN inventory issue for

November, what we actually need to do is quite counterintuitive. We need to increase marketing spend even further because we need to sell out of the CND inventory and it's not going to be profitable to do so. And so what we actually need to do here is we need to take me that's now inflated to 30% and we need to say hey you know we actually need to move this in the short term to 35%. And this additional 5% will be purely for driving all the CND inventory so that we can turn this back into cash

which is going to give us a more a better position from a cash perspective which is going to allow us to get back into the old revenue position. So let's then put it all together. First place to start is in forecasting. Forecasting is critical for anything financial related because it allows us to set our KPIs and our expectations that we're then measuring off on an ongoing basis to be able to call square. Anytime you are forecasting, the number one rule of forecasting in e-commerce direct to consumer is that you always need to forecast new customer

revenue separate to returning customer revenue. And the reason being is that both of these revenue buckets have different underlying levers that impact the realization of the revenue. So for new customers, this is primarily going to be marketing. Now, this might be just advertising spend for some brands. This might be advertising spend plus events and influencers. This might advertising plus influencers plus other tertiary channels as well. When it comes to returning customer revenue, yes, advertising spend is going to drive a little bit of returning customer revenue. Yes, an influencer activation might drive a little bit of

returning customer revenue, but primarily returning customers is going to come From uh product launches. It's going to come from marketing events like discount periods and it's going to come from other activations that are going to prop up and give returning customers a reason to come back. There is also obviously direct communication channels in here as well like email and SMS. These are different inputs compared to over here. Therefore, when we are forecasting, what are we doing? We're actually just forecasting inputs and therefore getting an output which is the forecast or the budget. And so we

need to look backwards into the inputs to be able to then determine how we're actually getting the final number. For returning customers, you want to look at all of your returning customer cohort. So every month back to let's go Jan 2021 all the way to Feb 2021. And this goes on and on and on all the way to today. There was a certain amount of new customers that you acquired. Maybe back here it was 100, then it was 110, then it was 150. All of these customers repeat at a certain rate over time. And you

can see this in your cohort analysis. And so people might come back at 7% in the first month, then 6%, then 4%, then two, then one, and then ultimately it asotopes down to usually a pretty small number. What you can then do is you can take all of these old cohorts and the amount of customers that you acquired then, and you can go, okay, well, they're about to enter into month 32 of them being a customer. What is the repeat rate on average of someone after month 32? And maybe it is 0.1%. So then you

take your 100, you times by 0.1%. You then times by whatever your average order value is on returning customers, which let's say it's $100. And so we would expect this cohort to give us $1 in returning customer revenue next month. And then we do that for the next cohort and the next cohort and the next cohort and every single cohort that we've had in the past. And this will then give us a realistic extrapolation of what we should expect returning customers to contribute to next month. This number ends up being usually within about 10% accuracy.

To get it within 1 to 2%, you do two things. Number one, you bake on seasonality. And so you look at uh average seasonality in the last 3 years across the calendar year and you just apply a factor based on returning customer revenue seasonality. Number one. Number two is you then go and superimpose marketing events that are going to substantially change these numbers. So obviously if you have a major marketing event that's going to occur this year that didn't occur last year, there's going to be revenue driven from that. you want to figure out what

is your uh expected revenue realization and then you add that into the forecast. Then on new customer revenue now in paid ads this is relatively straightforward when you have larger media makes or if you have other channels like influencers etc. This is where you have to build your own modeling around this and this is obviously where modeling and finance and data science becomes pretty critical once you once you achieve scale as an 8 to9 figure brand. To keep it simple for paid ads, what you want to do is you want to take your acquisition me

on the y- axis. You want to take spend on the x- axis and you want to plot the last let's for the sake of this video say 1 year of daily data. What you'll see is that every day there is a certain acquisition me that is associated to that day. And what should end up happening is you should have something like this. Now if you go and put a logarithmic regression or a linear regression whatever has the best fit for the model you will get an average of the expected efficiency at a certain spend level.

Now you will have outliers like These outliers over here. These are typically sales periods. So we want to actually remove these from the data set. We also know that if we're only going to be spending a minimum of $1,000 a month, we could also just remove anything under $1,000 a month out of the model too. And that's going to improve the accuracy of modeling on these higherend spends. Then what you're also going to have is all of these over here are going to be November and Black Friday periods that are significantly overinflating the actual efficiency

that you would expect during BAU. So you want to go and remove all these two. And then what you'll get is this line through all of the days last year at certain spend at certain efficiency. And then you know okay if we go and spend $8,000 a day here we know roughly what efficiency we should expect. And this is how you model out spend and new customer revenue expectations. You would actually back propagate from your new customer goal. So, if you wanted to get like, let's say, $100,000 in new customer revenue, uh, and let's say

you want to do it at a 4 a.m., you would just go to a four on the graph, which might be here. You would go across and you go, "Okay, it's right here. Can we spend enough to get a 4 AM and this much revenue?" And the answer might be no. Okay. Well, we fundamentally have an issue here. We need to rethink what we're going to do differently this year to generate outsized returns compared to the average of last year. Are we going to do some kind of marketing event? Is there going to be a

new product launch? Is there going to be a new channel? Are we tripling creative production? like what actual input or lever is going to generate this outcome. And this is ultimately the exercise of forecasting. The exercise of forecasting is to look at the realistic expectation against the target. There's going to be a delta. There's always going to be a delta. The board wants you to hit 50 million. You do this and you go, the math says we can only hit 40 million. And then you look at, well, what are the inputs required to achieve that

$10 million difference? We talked about the profit frontier. We talked about how funnels overattribute at the bottom, underattribute at the top. But I want to reinforce this idea cuz it's a really common mistake in budget allocation, which is that you will always see the best rorowaz at the bottom, you will always see the worst rorowaz at the top. And so if you have a meta campaign that's attributing at a 2x, and you have a Google campaign that's attributing at a 6x, don't simply go and put more budget here. This is going to be overattributing because

it sits at the bottom of funnel. It might have branded key terms. It might be retargeting. Even if you have a bunch of exclusions in place, probably bottom of funnel, and you're actually going to see better incremental impact putting budgets here. Obviously, always do this within a controlled test. Make sure that you're on top of whether this is actually true or not for your particular business. But be very careful with rorowaz reads in the platform because it's going to usually make you overallocate to bottom of funnel efforts and then you'll wonder why you're not growing

when you're increasing budgets. one thing that we haven't touched on at all here. And this starts to get a little bit outside of finance, but I think it's a really important mention, which is brand at 1 million to I would actually argue probably 20 million. You can just brute force revenue through performance marketing. Okay, performance marketing through ads can get you here very easily. You put $1 in, you get $4 out, and then you continue to scale. But eventually what usually ends up happening is you hit diminishing returns, which is that as you try to

put more spend into these platforms, as you try to start pushing past $30,000 a day, $40,000 a day in ad spend, you just can't get any further. And the way to go further is through some kind of brand effect. And I say this from personal experience myself. Obviously, in the early days, we worked with three 500 7 figure brands either full-time or in Some kind of consulting capacity. And these days, we currently work with over 60, eight, and nine figure brands and close to 10ig brands as well. And so we have seen both sides of

the spectrum. All these small brands that are very reliant on performance marketing and all of these large retail brands that aren't reliant on performance marketing at all. And in fact, there's actually a huge opportunity in performance marketing cuz they don't do it very well. And this is due to the brand that they have within the platform. You can open up and I do this all the time. You can open up one fashion ad account that's doing maybe $8 million a year and you can look at all the core metrics in the Facebook ad account, the

click-through rates, the CPMs, the CPCs, the conversion rate. And then I can go and open up another fashion ad account of a business doing $300 million a year. And the crazy thing is that the ad account over here has better metrics on everything. They have better CPMs. They have better clickthrough rates. They have better CPCs. They have better rows. Everything is better. And you go, how is that even possible? They're doing like 20x the volume. They're doing 20x the ad spend. That just doesn't make sense because as you scale paid media, you hit diminishing returns.

So why are they not seeing all their numbers degrade? And it's because of this overarching brand effect that they have in the market. They have so much market saturation. They have so many associations that have been built through external marketing efforts that sit outside of the ad account that inside the ad account it looks really good, but it's because of everything that they're doing outside of the ad account that makes it look good. And so I can go in any day on a brand that every single person knows and run ads and I'll have incredible

clickthrough rates cuz everyone knows who they are. But if I go in on a brand that no one knows who they are and I'm Trying to push, obviously the performance marketing has to be a lot better and that's why you hit diminishing returns. And so also just when you're thinking about marketing expense allocation, I would always be having some kind of budget towards branding efforts. And by branding efforts, it's making associations within the market that is going to put you in front of the customer where they are. And so if your customers are commonly in

a particular area or at a particular event or looking at particular things, that's where you want to show up to be able to build associations. I'll give you two personal examples of this, which is that I've recently bought running gear from two different brands. I bought from 2xU which is an Australian brand. It's actually a client of ours and then another brand which is 247 represent. And the reason I bought from this brand was because all of the running influencers that I follow, all the people that I watch YouTube videos of every week, all the

people that I follow on Instagram, they are all either sponsored by 247 or they just wear it. They make associations with it. And so because of that, that natural association that's been made within market of where I end up showing up on the content that I consume is the reason why I bought it. Had nothing to do with quality, had nothing to do with anything except for the fact that I follow all these running influencers, ended up following the founder, following his story, watching podcasts of him, and that's ultimately what pushed me to the purchase.

I would argue that all of those uh influencer deals, all of the branding exercises, them showing up to run clubs, etc., that probably doesn't have direct profitable ROI. They're probably not getting the coupon code that the influencer has at checkout, which I don't think they even do, but let's say they did do it. Probably not a profitable exchange, but it's the overarching branding effect of making those associations that ends up pushing Tons of people to purchase. On 2XU, it's the branding of premium. Now, 2XU's products are incredibly premium. I think they're probably one of the

highest quality products in Australia in this market. But honestly, I don't think that even matters for me in terms of my purchasing decision. I didn't purchase because I knew the product was quality cuz I bought online. I hadn't seen it. I bought because of the perception of quality and so it is the brand perception that they have built that this is the highest quality uh activewear clothing in Australia that is causing me to buy. Now once again was this through some kind of performance marketing ad? No. Was this through a Google ad? No. It was

through the overarching associations that they make. It's through the messaging that they have and it's through the way that they show up in the creative as well particularly in the campaign shoots that makes the perception that it is super high quality which ultimately drove me towards that conversion. And so both of these purchases likely wouldn't have happened through any kind of performance marketing effort. They actually occurred through brand which is why that we can't understate this and we need to have it as a portion of the video because it is unbelievably important particularly as you

continue to scale and it should be thought through as a budget allocation of an expense on the P&L. So wrapping this up, if there are five things that you should be walking away with as key takeaways to take forward in your business or working with a client, it is number one, know your gross margin and know how to calculate it correctly. You need to understand variable costs. You need to understand the difference between product margin and gross margin. You need to understand how that also then flows through into contribution margin. Number two is you need

to understand the definitions and you need to have live dashboards that track Acquisition me profit contribution ideally LTP to CAC not just being over here tracking rorowaz on a day-to-day basis and having weekly rorowaz reports this is not productive at all for core decision-m and moving the business forward the third is that you want to be separating all new versus returning customer metrics you want to be tracking new customer economics acquisition me new customer profit contribution new customer revenue, new customer cohort size separate from returning because ultimately the levers underlying them are different. This also

obviously applies into what we were just talking about around forecasting. Number four, you want to understand the difference between a cash verse a P&L play. It might very much so be the case within the business that the current limiter isn't marketing spend or marketing efficiency, but it's the cash conversion cycle. And so there is no point in arbitrarily pushing budgets up and trying to scale if it's just going to cause an increase in interest expenses on the P&L and a compression in me. There also needs to be constant communication between either the internal marketing team

or you and the agency as to the inventory position within the business across the different SKUs so that there can be strategies employed to be able to actually decrease profitability, decrease acquisition me, decrease efficiency, but prioritize the turnover of inventory into cash to make the business overall healthier. And then number five is that you want to be using the P&L and all of these other financial tools to be able to identify the constraint in the business. And so when you can understand how to read the P&L and structure it out, you can understand how to

KPI at each level. And then when You start falling below KPI, you can look above that level in the P&L to understand, okay, what has changed? What levers are there? And then how can we pull on those levers to rectify and course correct back to where the target actually is. If you made it this far, thanks for watching for an hour and a half. And if you are an e-commerce brand doing over $5 million a year, there'll be a link somewhere in the bio to reach there'll be a link somewhere below in the description to

reach out and get a free audit from ourselves where we'll run you through all of this financial modeling, but we'll actually apply it to your business. And if you're a performance marketer that's gotten this far, please reach out. We're always hiring for a play of performance marketers. Click on the website, reach out to us somehow. You could also email hiring bluesdigital.com.au and we'll look at your application. Most brands think they have an ad problem, but they actually have an offer problem. And most of the brands that think they have an offer problem actually have a

discounting problem. So, in the next 90 minutes, we're going to run you through the why as to why offers matter so much in e-commerce. What an offer actually is if we break it down into its fundamental components. We'll then go through the economics of crafting an offer that works for you. We'll go through the five mechanics of an offer in e-commerce. I'll then show you the right offer. How do you make an offer that actually fits your particular brand? Because it changes niche to niche. Then we'll go into offer discipline. Lastly, I'll give you the

seven offer failures that you don't want to do. And then we'll tie it all together so you know exactly how you can improve your offer in your store right now to see better performance across the entire funnel. Now, the reason why offers are so important is because they are the single biggest Change that a DTOC operator can make to have the largest inflection upwards in performance with practically zero cost. We're not saying go and make more creative. We're not saying go and design an entire new website. We're saying if you just change the offer, the

way that the value is presented to the customer, you can generate significantly higher conversion rates, higher average order value on the exact same cost to acquire a customer. Most marketing calendars for 8 to 9 figure brands are just 52 weeks of weekly promotions. And the result ends up being that the customer is just trained for discounts and they think that they have offers sorted because they're rotating in new offers every week. But that's the complete wrong definition of how you should be thinking about this problem. A 25% discount on a $100 average order value erodess

gross margin from $60 down to $35, which is nearly a 50% reduction. Now, even if average order value lifts by 20%, you're still in a worse position. So, let me show you what an offer actually is. Because if you can't explain exactly how an offer works, you definitely can't design one. So, this is the hierarchy of influences to how you can think about all of these different variables as they pertain to performance within the business. So, when you zoom out and look at a direct to consumer or retail business, the least important thing, but it

builds the base that allows everything else is the budget. Ultimately, if you have more budget, you will sell more product. If you have less budget, you will sell less product. And so, the budget that you actually put into the platforms is effectively the fuel on the fire that allows everything else to go. Now, if everything else is bad, it won't do much. Which is why the next step is account structure. Now, you Can put all the budget that you want into the funnel, but if your account structure is set up in a way that's just

retargeting the same people over and over again, or it's pushing the wrong products, or it's got a myriad of issues with it, then it will let down everything above it. But once again, these two things are at the base of importance. These actually aren't the important things that are going to drive substantial delta in the business. Then we move up and we go to creative. Substantially more important than account structure and budget. This is how people actually hear about us. Then we get to the offer. This is what they're actually getting and why they should

purchase right now. Then the brand is why people should buy from you rather than a competitor. And then the product is why people buy it all. And this becomes incredibly important at super bottom a funnel in repeat purchasing. So this is really the hierarchy of importance when it comes to creating a purchase decision within the consumer. The higher up the pyramid, the bigger the lever, but the harder it is to change. And because it's hard to change everything up here, most teams just spend all their time down here changing account structure and budgets around thinking

it's going to do something. The top three is actually where massive delta becomes unlocked within performance. An offer doesn't just mean a discount. That's not what we're talking about here. In fact, you don't have to discount at all and you can have a unique offer. And we'll give you a bunch of examples of that later on. A discount is a percentage off or a coupon code or a store-wide markdown. Whereas an offer is product positioning. So, what problem does this product actually solve for who and why right now? Number two is the architecture of the

price. So, it's not just are we Doing a percentage off, but we have more optionality here. Are we doing some kind of threshold offer? Are we doing a units per transaction offer? Do we have a bundle structure where we're discounting by 20% but we're forcing units per transaction to be up at three to four so we actually don't take any gross margin compression. Number three is the value mechanic. Ultimately when it comes to an offer what we are doing is we are creating a price discrepancy within the market. And this, in my opinion, is the

best way to think through this problem of creating value and generating an offer that converts people at higher rates, which is that from the consumer's perspective, the value of the offer is the perception of the gap between the cost of goods sold for the business and the value that they're actually getting. That's why a blank 50% off or 60% off discount works so well, because a consumer sees that discount and they go, "That's got to be a cost of goods sold. this business can't even be making money on selling that. Hence, that is a fantastic

deal because no one is making money in this exchange. So, I am getting all the value arbitrage. However, from the business's perspective, a good offer is the opposite. It is how do we get people to spend as much money as possible above cost of goods sold whilst making them think that they're paying cost of goods sold. And so, a consumer wants to know that they're getting a good deal. They want to think they're getting all of this value and it's only costing them this price. Wow, they must not be making much money. But then the

advertiser or the business selling the product is actually making a ton of money. But it's the way that they have reorientated value within the offer that Is making the consumer think it's a good deal. And sometimes it is just a good deal for both. Like sometimes the company might own a software company and therefore they can give free licensing deals away to some software with the product. And so for the consumer, it is an incredible deal because maybe they get to save on their tech stack when they get a software. And from the business's perspective,

it's also a great deal because they own the software anyway and it costs them absolutely nothing to deliver it to the customers as an add-on. And so it's great for them too. And so the value can obviously coexist for both the advertiser and the consumer at the same time. And that's what creates an incredibly valuable offer. So on the value mechanic, this is how we're creating the perception of value and effectively collapsing the consumer towards the perceived cost of goods price. So this could be free shipping, free gift, buy one get one, samples, second order

tier, a subscription unlock. There's a lot of different mechanics here that allow us to increase perceived value. The next one is risk reversals. This exists in every industry. It's not ecom specific. This is where you're going to reverse risk by having a returns or a guarantee in place. Maybe there's a trial period. Maybe it's a sample offer, so you're just buying the sample and that's it. And then there's urgency, which has existed for as long as time in marketing, which is that you want scarcity. If it's a drop, if it's a seasonality tie-in, if it's

just a window in which this product's available before it goes out of stock, this is your fifth lever to be thinking about. So discounting, technically, sure, is an offer, but it's only one of five options In terms of how we actually create value within the offer. It's only really correlated with price architecture. Everything else, how we position the product, the value mechanic, the risk reversal, the urgency, all of this should also be used in conjunction with the discount or the discount doesn't even have to be applied at all so that we can create the best

offer possible. Now, some of the best offers that exist do not use flat discounting. Flat discounting is really easy math to run. It also substantially impacts uh your perception in market of where your price is anchored. And so if you end up flat discounting too much, well guess what? All the customers get trained that you're on flat discount and then you become a discount orientated brand and now your margin permanently compresses. You launch the business thinking you were at 60% GP. Now suddenly you're at 45% because you always have to discount because it's the only

way that people come back and buy again. And so before you go into any kind of offer curation, it's really important that you understand the economics of discounting. But let's run through a base example. We've got a $100 average order value. We've got a cost of goods of $40. For the sake of this, let's just assume cost of goods has all variable expenses, shipping and fulfillment, transaction fees, etc. Let's just assume it's all in there. Gross margin or gross profit is therefore $60. Our break even rorowaz is 1 divided by our gross margin. Now, our

gross margin here is 60%. Therefore, our break even rorowaz is a 1.67. Now, our Target CPA at a six rorowaz is average order value divided by 6. So 100 / 6 = 1667. And therefore at this target our contribution profit per order is 60 - 16 which is $43.33. So this is the brand at full price. Every single brand should know this exact waterfall for every single product in the business. Then what we need to do is model out the offer. Now we'll start with a very simple offer which is not what I would run.

I think it's pretty bad, but 25% off sale. So, what's going to happen here? Well, average order value, or at least let's call it price per unit for the moment, and you'll understand why we're changing this up in a second. Price per unit compresses to $75. Now, here's what people sometimes don't take into consideration when they do this modeling, which is that when you do a discount, units per transaction increases. So, units per transaction is the amount of units in the cart in the order. Let's say that in this case on average there was one

unit per cart which is unlikely pretty much everyone has above an average of one because someone is going to add two things to cart and it's going to pull this up. For the sake of simplicity let's say that this is at a one. Now once you go on sale people buy more because people want to capitalize on the sale and so units per transaction might actually go up substantially. Maybe you never go on sale and so it rises all the way to a 1.63. Now what that means is that average order value which is price

per Unit or average unit retail times by units per transaction equals $122. So actually where we thought discounting would compress average order value, it actually increased it. We actually get more money on each customer. However, our cost of goods sold is still 40% but it's not 40% of this average order value. It's 40% of the retail value which is $162. Right? This is $162 retail price prior to us discounting by 25%. So our cost of goods is this times 40% which is $65 which equals a gross margin of $57. Our break even rorowaz therefore increases

to 2.22. And then if we want to hit the same contribution profit per order that we were outside of the sale. The way that we calculate this is we take the gross margin, we minus off our target contribution margin, and that gives us $13 as a target CAC or cost to acquire a customer. If you want to convert this to return on ad spend, you just take average order value and you divide by CAC, which is 122 divided by 13.67, which is an 8.97. So coming up on a 9. Now, the rorowaz previously was a

six. So, we need to increase return on ad spend by 50%. Okay, we need to go from a six efficiency to a 9 efficiency on a 25% sale off campaign. And this is why understanding the economics of an offer is so important because on the surface when you just run the topline numbers, it looks really good. Okay, if we run this discount and units per transaction go up, we're going to make way more money on each customer. We're going to make $122 rather than 100. This is great. The sale is working in our favor. We're

doing more revenue per customer because of the sale. But then once we start to waterfall down through gross margin, which is actually lower. So even though we're doing more revenue here, we're doing less gross margin. So gross margin as a percentage is compressed enormously. Then once we go to the break even row, it lifts quite a lot. Then we go to the CPA and it kind of gets ridiculous. And then we go to the rorowaz and we go, wait a second, for us to hold the same level of profitability, we need to be 50% more

efficient. Is that going to realistically happen? Now for some people, yes. I know a ton of clients that come to the top of my head where we actually model out 70 to 80% improvements in efficiency during sales periods because they never go on sale and they're a luxury brand. And so when they do, the spike in sales is enormous. I have other brands that I can think of where we would model in a 5% improvement in efficiency because everyone's just so used to them discounting that it doesn't matter to them and they have so much

market saturation and penetration into the Australian market being known as a discount brand that it doesn't matter if they discount doesn't really lift sales that much materially and so you really need to understand how your efficiency reacts to different levels of sales so that you can model out what you believe efficiency lift will be to understand whether the offer is actually going to work for you or not. Now, I'm going to get ahead of myself a little bit and I'm going to run you through a way better offer and all of the economic mechanics behind

it so you can understand why offer curation becomes so important. Now, a lot of what I'm about to throw onto the board, I haven't even spoken about yet. It's further in the video in terms of agenda, but I'm just going to throw the whole kitchen sink at an offer here so you can get an understanding of how you can go from a 25% offer to instead something crazy with so much free stuff with so much layered in and you actually have better gross margin. So, let's actually contextualize this business and let's call it a supplement

business. Here's the new offer. So, the offer now is buy a 90-day supply, get 25% off. You get free access to the brand's app where they have meal plans for you, dedicated meal plans every single day. On top of that, you get a free mystery gift. Don't know what it is, but you get a mystery gift. And you also get this 25% off discount for life as long as you stay on subscription. And normally, the subscription discount is 10%. And so you're getting double the normal subscription discount as long as you stay on subscription and

you don't cancel. So what does that do to the mechanics? Well, on average order value, it's going to increase average order value by about 2 1/2x. Reason being is there's already a bit of take rate on the 90-day offer up front. So this isn't the 90-day offer. 30-day might be about $80. 90-day is $21. And so there's a little bit of take rate here, which is pulling the average order value up, but it isn't great. Now, average order value jumps to about $190 because the take rate on this shoots through the roof. Now, how is

the actual margin profile impacted? Well, cost of goods is 40% and retail price here is 253. Spibble are getting a massive discount. And so, cost of goods is 101 based on 40% cost of goods. However, cost of goods isn't 40%. And the reason being is that when we ship three of the product, we actually get economies of scale in shipping and fulfillment. So where on shipping one unit it costs us $12 to ship the package. When we ship two units it only costs us 14. And when we ship three units it only costs us 17.

Now we have the unit cost baked into the price as it scales. And so we are technically charging onto the customer 3x this amount baked into the price the Base price of the product. But we don't have to pay this amount times three. We actually get a quite a large saving here. We've baked in $36 into our pricing for shipping. it's only costing us 17, which means we actually have $19 in savings we've just pulled together. So, this cost of goods or cost of delivery actually drops to 82, meaning our gross margin is now $108.

Now, if we want the same contribution margin per order, what do we do? We just minus this off. To get this contribution profit per order, we need a cost to acquire of 65. 190 / 65 equals 2.9. So to make the same contribution margin per order now we only need a 2.9 rorowaz rather than a six. Our efficiency can have and this is ultimately where evergreen acquisition offers like this crush. This isn't an offer that you rotate in on Black Friday. This is an offer you could run all year round because it substantially compresses the

efficiency that you actually have to sit at. Now obviously we probably don't want to go down here. We just want to increase contribution margin per order. Rather than making $43 per order, why don't we make 50? Why don't we make 60? However, we did some other things in this offer, too. We added a free mystery gift. Now, we didn't actually include that in COGS. The reason why we didn't is it's inconsequential. You can put free mystery gifts in that have a $2 to $3 cost of goods. And you can get pretty flexible here. You could

just give sample packs from other products. If you're in supplements, um you could just throw in something like a drink bottle and as long as you make a very large PO, The cost of goods on that drink bottle throwing can be very low as long as it fits within the package and doesn't drive up your shipping and fulfillment costs. There's a lot of stuff that you can do on mystery gifts. You can do mystery gifts that don't even have a cost of goods associated with them, right? And so you can do a partnership or a

collaboration with another brand in which another brand actually gives you the gift to give away and it's co-arketing because the other brand is like, if we can get some free samples in your orders and you can frame it as a free mystery gift, amazing because these people might come and buy from us. And so there is tons of stuff you can actually do here to be able to have a $0 cost of goods or a1 to $3 on a mystery gift. Then we have the lifetime subscription discount. Now the objective of this is to keep

retention high and keep them on quarterly billing. And so what will happen is to get this 25% off discount, they not only have to buy a 90-day supply, but they have to tick the subscription box. If they don't tick subscription, they don't get the offer. Now because of that, it pulls people in to the subscription. Number two, they get a discount that they otherwise wouldn't get if they subscribe at a different point of the year. So, you can put urgency around lifetime subscription discount. And then what that does is substantially increases LTV, particularly on a

90-day basis because on the 90-day is when all of the repeat transactions occur and substantially bumps up cohort lift. Not only can you be more aggressive on acquisition here to acquire more customers just on first purchase contribution profit, but you could probably be even more aggressive because the actual retention on this offer versus this offer is going to be substantially different. You're probably going to have double the amount of repeat purchasing coming through from This offer than you will over here. And then you can continue to stack really lowcost things into the offer here. What

you do have to weigh up is it can get to the point where you're just stacking so much stuff that it's it's just too much for the consumer. Simple scales when it comes to offers. Obviously, you want to layer in as much meaningful value as you can, but you also want to sub subtract as much as possible to only give the stuff that people care about. And so a really valuable exercise after this offer would be live for let's say 90 days, 120 days is to then survey customers and get an idea of which of

these three additions in the offer did they like the most that convince them to buy. And you might find that no one could care less about the mystery gift. It means nothing. And therefore, let's remove it and let's test adding something else or let's just not have it there at all because it just adds additional complexity and let's see if conversion rates are impacted. You might find that the app really got a lot of people to buy and this really pulled them in. But then once they went onto the app, they were like, "Oh, this

is useless. This is just a bunch of chatb meal plans. I don't really care about this at all." And it decreased brand perception and impacted retention revenue. And so even though this helped a lot on the front end of the offer, it actually substantially negatively impact the back end of the offer. And so then we make a change here. We either change the product and the deliverable or we pull it out of the offer entirely. And so this is where you iterate on an offer over time to improve the acquisition economics. So you're making more

profit on first purchase, you're driving down the cost to acquire a customer, but you're also working to improve retention, too. There's two curves that you really need to understand here when it comes to the economics of an offer, which is that as you increase flat discount rate across the offer, your Required return on ad spend to break even increases linearly. And so your break even return on ad spend at a 10% might be 1.9 and then it goes up to 2 and then it goes up to 3 4 5 etc. And then eventually obviously the

break even point becomes pretty much zero because sorry return on ad spend reaches infinity because there is a point in which there's no margin left. This is fine. I think most people have their head around this. What they don't have their head around is that target return on ad spend increases exponentially. And so if you have a target rorowaz goal based on a contribution profit target. So as we went through before in the example, let's say that your goal is to hit $40 in contribution profit. The reality is is that margin compresses to $40 much

faster than it compresses to $0. And so as a product of that, if you were to graph this, and we have tooling that allows us to do this. So reach out to us if you're a brand and we'll just send it through and you can model all this yourself, which is that your return on ad spend will actually look like this. And so as you increase discounting your required rorowaz to hold the same contribution profit increases exponentially which means that the difference between like let's say a 15% discount and a 25% discount can be absolutely

enormous in the rorowaz requirement. This might require a 20 rorowaz. This might require an eight. And it's like wow we're never going to hit a 20. So any kind of discounting at this range of the graph is just inapplicable for us. We can't even do it. This is another graph that's really important to understand. And if you can model this based on your own data, even better. Now, we will do this modeling sometimes for brands. It just depends on how clean their data set is over time of different flat percentage discounting so that we can

see historical units per transaction change. So, what we're Really looking at here is as we move across the x-axis, discount rate is increasing. So, we're going from 10% to 20%, 30, 40, and then let's go 50 all the way over here. On the y- axis, we have contribution profit per order. Now, what's going to happen is as you go and discount by like 5 to 10%. You're actually going to make less money. And the reason why you'll make less money is because this is such an unconvincing offer, no one really cares about a 5% discount

or a 10% discount that it's not going to increase people's units per transaction. People aren't going to put more items in cart because of a 5% discount. It's meaningless to most people. And so as a consequence of that, your average unit retail or your unit price will just compress by 5%. And there'll be no upside. And so because of that, your contribution profit actually declines when you have these really small discounts. Then at some point there's an inflection and people start caring about the discount size. And at that point, your units per transaction starts to

increase. And so if we look at average unit retail on this graph, this is what average unit retail looks like because you were just discounting the prices of all your products. So the price is going down, but the important component becomes units per transaction. Now units per transaction stays the same here. And then there is a point in which people start to care about the discount and it goes up and then the reality is as the discount gets more and more aggressive, it continues to go up. But if the compression in margin doesn't outweigh the

increase in units per transaction, guess what? Total contribution profit declines. And so it is the function of units per transaction and average unit retail that will produce the contribution profit result. You want to understand how aggressively does up increase as we increase the discount threshold. And then obviously you just get to a point in discounting past like 30 40%. in which it's just not worth it for anyone because no matter how much units per transaction increases, you're making like no money because you just compress margin down. Now, why does this theoretical graph matter at all?

Well, because what you want to ideally be doing when you're thinking about what kind of discount are we at least surfacing within the offer. Now, it doesn't have to be a discount offer, but what I showed you before was a really well-curated offer that did still have a percentage off that was headlined. You want that percentage off to sit where contribution profit per order is maximized. At what point do we give away some margin, but we maximize the average order value lift? And it is at this point that we want a discount. And for this

brand, this would be at 20%. Once again, you can model this out yourself as long as you have some historical data points on units per transaction during different flat discounting periods. And that just depends on the consistency of flat discounting offers that you've done over time. So, let's run through the five different offer mechanics. Number one is product repositioning. What it changes is the persona, the problem, or the headline. It's the exact same skew. So we're not actually changing the product fundamentally. We're not doing any kind of bundle, but we are repositioning the product. The

contribution margin impact is nothing. In fact, sometimes it can actually be positive. And it's best for doing it on one skew or doing it on a hero skew. So there's no better example to do this in than on a supplements brand. So let's go and put Groans. Now what you could do, and what average marketers would approach this brand with, and for those that don't know, they sell Daily Greens gummies. Average marketers will position the product like this. Daily Greens gummies get your vitamins and that's the core angle. That's the way that the product's positioned.

But instead, what Gruns does very effectively is they actually reposition the product into almost six different products that serve six different audiences and play to six different mechanics or unique mechanisms. Number one is gut health for bloating. Number two is GLP1 support. Number three is fiber for digestion. Number four, hair health. Number five, cognition dash focus. Number six, multivitamin for business professionals. And they have way more than this. What we're really doing here is we're just listing out different personas and concepts that they're testing and the core ones that work the best in the account.

But persona testing also flows directly into the offer because we create offers around high performing personas and we can craft the product title, the product position, every other component of the offer around this. So, as an example, and I don't know the compliance on this, but I'm going to throw out an idea, which is that let's say the GLP1 support is one of their highest performing angles, and they're driving a few million dollars a month in new customer revenue through this angle. or they could then go and craft an offer specifically for this angle on

a dedicated landing page which has something along the lines of the GLP1 angle going into it repositioning maybe the product title slightly and then if you hop on a subscription on a 90-day billing cadence so we collect 90 days right up front to be able to improve the profitability of acquisition then you also get 25% off your GLP1s with our registered partner and so you would Partner with an actual GLP1 provider You would ask them if you can give 25% off to your customers that flow through from this funnel and then you give this away

to them. Now, cost of goods here, $0. In fact, you could actually probably make money on this offer because you asked the GP1 provider to give you a 25% off discount and then a 20% clip on all future purchases or something like that. And so, you would actually substantially increase the profitability of this whilst not changing cost of goods and just giving something away for free. And that's how you took the product, you repositioned it, and then you crafted the offer around it. Now, you could do this for obviously all of them. So, if we're

repositioning the product into hair health, as an example, one really important piece of daily greens impacting hair health might be that really when you do double blind double blind placebo experiments, the hair health impact is only measurable post 90 days or post 120 days. And so you make that unbelievably obvious in the front- end offer. And so you go, "This is an incredible product for hair health, but it does take 120 days of consistent usage." And so because of that, and because our studies have shown that, because our customers say that, here's all the testimonials,

we're giving you a special offer to make sure that you can achieve this goal, which is that our normal 90-day subscription, we're going to throw in a free extra 30 days in your offer. Now, how are we actually mechanically going to do this to be able to preserve margin as a business? We're not going to give you four in the first order. What we're going to do is we're going to give you the three, but on your second order, you get one free. So on The second billing, when they get charged for another 90, which

takes them to 180 days, one of these within the second order is obviously discounted 100%. Which is giving them like a 33% discount on the second purchase. And so that's where you're using the product repositioning and you're building it into the offer for congruency. Now, none of these, and we could go all day. I could come up with 20 different offers for each individual product persona here. Really, the key thing to understand here is none of this would have been possible. None of these offers could have been made unless we started repositioning the product. Cuz

if we just kept it here and we removed this and we just go, "Okay, it's a daily greens gummy. What can we do?" We're incredibly restricted and limited into what we can do here. Yeah, we could, sure, we could add on 25% off GLP1s, but this is irrelevant to 90% of people that are buying the product, right? 90% of people aren't on GLP1s. Or maybe they are. I don't know. And then the same for hair health. 90% of people probably aren't buying for hair health. And so it completely cuts the ability to do any kind

of offer around the clinical studies here and substantially restricts us to just very broad, very generic offers that are not going to cut through on a specific addressable market. All right, the next offer mechanic is using a bundle. Now, what it changes is units per transaction goes up obviously because you're forcing people into buying multiple products. You get cost of goods efficiency specifically in the shipping and fulfillment, not in the actual cost of goods. And then contribution margin should positively be impacted if you craft the bundle correctly. And then this is obviously best for multisq

brands. Unless you can package one product multiple times like we did before with the 90-day trials. If it's a consumable, if you don't sell a consumable and you only have one Product, you can't really do this. Now, I really don't think I have to explain bundles. I already went through economics and discount mechanics across discounting. So, the exact same same thing applies here. Now, bundles isn't just taking different SKs and putting them together, but it can just be the same skew given away twice. And so, if you sell board games for kids, buy two, get

25% off. Like, this is a very common offer that everyone's very familiar with. This is probably what everyone thinks when they think offers. In fact, one of our interview questions that we ask is, "Explain to me what a good offer actually is." The most common answer is just high discount percentages or some kind of bundle because it is advantageous to the advertiser because they protect contribution margin which is correct. The issue with bundles is that you could do a lot more and so a bundle is pretty simple. It's an easy offer. It makes sense as

long as you do the economics correctly, but you can take it a lot further and be a lot more clever with how you're putting these offers together. The next mechanic, which is one of my favorite to be honest, is gift with purchase. What it changes is up if you force units per per transaction up with the gift with purchase with pretty much no change in contribution margin. So there's a zero impact. And this is for anyone that has a secondary skew that they can give away for free that's incredibly low cost and pairs well. I'll

show you the economics of why I actually like this so much. So let's say you sell towels, beach towels. When someone buys one, it's $50. When someone buys two, it's $100. But when they buy two, they get a free beach bag towel. Sorry, they get a free bag that holds the towels as well. And then because beach towels might be bought by families, you could probably do a buy four offer as well for families of four. And you could give a further tiered giveaway, but let's just keep it simple. Now, the cogs on the tow

is $15 a unit. So, it's 15 over here, it's 30 here. Now, the cogs on the bag is $0 here cuz they're not getting a bag on this one. They are. So, it adds to our cost of goods, but it's super cheap. It's $4. Now, shipping and fulfillment fees plus 3PL is $850 here on one unit. Over here, it's $1250. Which means our contribution margin is $26.50 and $53.50. Meaning, if we do a break even calculation on rorowaz here, the break even rorowaz on just selling the towel is 1.89. The break even rorowaz on this

offer is 1.87. So, contribution margin almost doubled. In fact, it did double. It over doubled. And our break even return on ads, the efficiency that we have to operate at, decreased. So, we can be at a worse efficiency and make double the amount of money, which is a crazy position to be in. And this is why gift with purchase offer can work so incredibly well as long as it is convincing. Now, this offer, the economics of it look incredible. Where it falls apart is do people really care about getting a bag? And that's why the

offer actually has to be well thought through. You have to understand the customer correctly and you have to be giving something that has a high perceived value. Now, how would You make the perceived value of this bag increase? Well, you would actually sell it on the website and you would sell it for like $60. And so then when you present this offer, you don't present it as $100 plus you get a free bag. You present it as $160 has been discounted to 100. And so you're getting like 40% off by buying two. What's even better

about this is what I showed before, which is that you can implement gift with purchases in conjunction with other mechanics in the offer. So, this doesn't just have to be the offer. You can layer more stuff in. And so, on top of this, you would take whatever the biggest objection is or whatever the feedback is from customers. And by the way, the offer always starts at the customer. It doesn't start at you. And this is where people really get offers wrong is they go, "How do I make the most amount of money? How do I

set this thing up so my contribution margin and break even rorowaz is as low as possible and as high as possible. If you do that, you'll just fail because nobody cares what you want. It matters on what the customer wants. And so when we then go and add another variable to this offer, we're not just going to arbitrarily throw something in like, oh, let's just tack on like a prize pool that you can win like a holiday to Fiji that's worth $5,000 and then we'll give this away once a quarter. And so there's also this

perception of you could win a holiday when you buy this offer as well. Cool. That's great. But does our target demographic care at all about going into a prize pool for some holiday with no known nods and no no information about it at all? Maybe, maybe not. And so this actually starts with understanding the customer and directly surveying them and asking them, "What would you like to see from us? What held you back from purchasing originally? What was the concerns when you first bought the product?" And you might find out, well, the concern was that

we actually thought the towels would be relatively low Quality and that they wouldn't last, particularly for kids in the family who have ruined towels in the past. It's like, okay, great. Well, when you buy two, we're going to put in a lifetime guarantee. And so, if anything ever goes wrong with the towel, we'll send you another one. Now, if that is the biggest objection from people buying and you then place that in the higher average order value offer, well, guess what? Not only you probably going to increase conversion rates, but you're going to increase conversion

rates on this offer, not this one. So, you're going to give people even more reason to go up and spend more. And then once again, this for you might not matter at all. And this is why offers need to be curated to you specifically because a lifetime guarantee on some businesses means absolutely nothing. Nobody cares about a lifetime guarantee. If I'm buying some creatine gummies, I'm going to eat them and then they're done. And so, we need the lifetime guarantee or we need whatever the offer is to actually meet the customer where they are. So

the next mechanic is tiered offers. This increases up because we're saying that as you spend more, you get something. It generally improves contribution margin if it's structured correctly. And this is best for people that have a binomial average order value distribution. Now what that means is that if we take average order value and we put it on the y ais, if we take average order value and we look at all the different orders within the business and we put these across here, then the amount of customers that are buying. So you could do this as

a bar chart. So people that are purchasing $0 to $5, this is how many people, this is how many people, etc. I'm going to do it as a line graph cuz it's a little bit easier for me, which is that average order value will generally look something like this for a lot of brands, which is that there is a point in which most people buy within this average order value range and then There's some people up here and there's some people down here. Now, for quite a few brands as well, you will have a binomial

average order value distribution. And this could be based on the actual current offer structure which is that people can buy one or they can buy three or this is very common when you sell a lot of products. If there's a very large skew count like in fashion, people can kind of buy whatever they want. They can put their cart together in any way. And so what you'll end up seeing is there'll be a peak and then there'll be another peak. This is called a binomial distribution. There is two peaks that exist within the distribution set.

So where does this start to cause issues? Well, if we look at average order value in Shopify, average order value might say that it's right here. This is your AOV. Now, the issue is if you go and set an offer based on this average order value number, it's not really going to do much because let's say that you set an offer, this average order value is at $100 and you want to do some offer at $ 110 to try to push people up. So, maybe it's you get free shipping at $110. If you set free

shipping as an offer here at $110, well, it's not convincing any of these people to spend more because they're already spending over $110. So now you're just giving them free shipping for free and you're just losing margin on them. And these people aren't going to be convinced to go up to $110 because they only sit at like 70. So you're effectively telling them to almost double their average order value to get free shipping. They're not going to double their average order value. And so the only people that you're convincing to spend more is just really

these people right here. And this is a tiny fraction of the total amount of customers. This is maybe 10% of customers. And so you've just given away enormous amounts of margin on all of these users, all of these customers. And Then none of these people are getting convinced to actually spend more. So it's a bad offer. Hence why if you have a binomial distribution, you need to instead look at the distribution curve, not look at any averages, and then build an offer around the two peaks. And so in this case, we would want to set

an offer somewhere here that's targeted at these guys. And then we would want to set an offer somewhere here targeted at these guys. And the idea is we want to shift both of these up. Now once again, you don't have to do this in isolation. This doesn't have to be your whole offer structure. You can use all of these mechanics together. And so you can have the gift with purchase to incentivize one of those peaks up. And then you can have a second offer for the second peak that uses a different mechanic that we've gone

through. Then you can also just straight discount. And the idea here is that it will decrease your cost to acquire a customer. the contribution margin impact will probably be negative unless uptweighs the margin erosion which obviously you could do this correctly. I showed you the graph before where you want to find that point in which units per transaction increases and it maximizes your contribution margin. So if you do this right you actually do make more money. If you do it wrong you make less. And this is best for two scenarios. Either you have a few

SKs that have incredibly high gross margin and then you just discount these or you have grade C inventory. to inventory that just isn't moving and you need to get rid of it and you need to turn it back into cash and you focus the discounting here so that you can turn it over. Now, to do this well, you want to do a couple things. Number one, you want to model average order value lift based on previous discounts and how they have impacted average order value and units per transaction. That way, you don't over discount and

just lose a bunch of money, but you discount to the correct amount in which you maximize AOV and maximize profit contribution. Then you Also model out the target return on ad spend needed at this new discount threshold. So not only do we know how much average order value we should expect, we also know what does our target efficiency need to be on the platforms to be able to hit the profit contribution targets that we're after and is this reasonable. Can we actually hit these return on ad spend, these efficiency numbers? Number three is you want

to do this very rarely, one to two times a year. Flat discounting is obviously going to train the entire customer base and anyone that's warmer in market to the fact that you discount all the time. So you probably want to preserve brand equity and perception in market. And then number four is that you're using it to turn stock into cash through products that aren't moving as opposed to just trying to rescue topline revenue. These are just fundamentally bad businesses which is the ones that have to discount to achieve a topline revenue figure. They'll always have

worse margins. They'll always be in a worse position. and the cash flow is a lot worse. Unfortunately, this is most businesses though. Cross- sales and upsells sit out separately from the five offer mechanics. And the reason being is that they sit downstream from the primary offer. The main thing you always want to be calculating and modeling out when it comes to a crossell and an upsell is the take rate. And what we're really optimizing for here is incremental contribution profit per order which equals your take rate percentage times by the average order value that is

going to be associated with what is ever what is being taken and then contribution margin or gross margin percentage. So the take rate percentage is the number of customers that are exposed to the offer and then actually take it. The average order Value for the upsell or crossell is the dollar value of whatever it is that they're adding to cart. So, if you're saying, "Hey, here's an extra $20 item you can add." This would be $20. And then the gross margin of is obviously whatever the gross margin is on that product. And so, if we

want to run through a really clean example here, let's say 10% of people take the upsell. The upsell is $50 and the gross margin is 50%. Then we have $2.50 in incremental CP. Now, this doesn't seem like a lot, right? You look at this and you're like, "Ah, this isn't really even worth it." But it is because at the stage of the crossell or the upsell, the order's already locked in. Okay? They've already locked. They've already decided on the primary purchase that they're going to make. And at that point, right before they're about to put

in their card details or often postcard details on the actual post purchase page, you give them an offer. You say, "Do you want to add this in as well?" And then this is just free additional incremental profit if there's a high take rate and if it's a good offer in the first place. Let's say that you were actually only making $20 in contribution profit on acquisition. Well, now this upsell increases that by over 10%. Which may actually allow you to spend more on acquisition. Now your CAT can be $2 higher to acquire even more customers

and take more market share, which allows you to scale faster. Now, where people make a massive mistake on the cross-ell or the upsell, is that they focus on the margin of the product. Let's push our highest margin product as the upsell because that's what will make us the most amount of money. It's not true. So let's say you have upsell A which is low margin and then you have upsell B which is high margin. Now on the low margin Product because it is more congruent with what the customer actually bought. The upsell makes more sense.

It's potentially just a better offer. Well then because of that you have a 20% take rate. But on the high margin products people don't want this as much. is not as good of a fit and therefore you have a 10% take rate. Average order value lift here is $40. Average order value lift here is $30. Gross margin here is 60%. Gross margin here is 30%. So gross margin here is terrible compared to this. So what you need to understand though is take rate isn't everything. The thing that you need to be optimizing for on an

upsell is the incremental CP per order. So let me run you through an example. Let's say that you have upsell A, which is low margin, and upsell B, which is high margin. On upsell A, we might have a way better take rate. Okay, we have a 20% take rate compared to 10%. Double the amount of people are taking us. Our average order value lift is also higher. We're making $40 extra rather than 30. But on upsell A, we are low margin. We're only 30% gross margin. And that's maybe because we're giving away a pretty heavy

discount here. And also, it's just a low margin product. On upsell B, we have high margin. We're not discounting it. We're giving it away at the same price. Hence why the take rate is way lower and hence why average order value lift is way lower as well. Then when we go and times all of these together, which is what the formula is up the top here, what do we get? We get a $160 over here, we get a $180 over here. So we actually make more money on the half take rate upsell with lower average

order value lift because it had better margin. And so don't just blindly optimize your upsells and cross sales Based on take rate or average order value or gross margin. Any of them independently doesn't matter. It's how they all flow into the mix to drive incremental contribution profit. Quick side note on exit intent offers as well. Whether this is you're doing the popup where someone goes to actually exit the browser, it gives you a pop-up, which I don't love. I'm not a huge fan of those. They're pretty gimmicky. But the other option is just an abandoned

flow uh that you can set up via emails as well. There's three different stages of exit intent. Someone can exit just when they're browsing. Someone can exit when they've already added something to the basket or someone can exit when they've gone all the way to checkout. What it tells you is what stage in the funnel they're actually at. So if they're browsing and they exit, they have no commitment. They haven't shown that they actually want to buy anything. They're actually still probably at the level of product aware, not most aware. As they have actually added

something to basket, they've showed that they want to buy and as someone's reached checkout, they pretty much are there at the purchase, but there was some kind of friction that stopped them. Potentially, they just need a reminder to go back and buy or they need some kind of risk reversal or change in shipping offer. So, when you think about the offer logic here, you don't want to treat these customers the same. You want to split out the offer accordingly. So, if someone's just browsing, you want to be generous. You want to actually get them back.

You want to give them a reason to buy. You want to put urgency associated with it. If someone has added something to basket, you don't want to go and give away all your margin. This person might just be taking some time and they'll buy in the next one to two days. So yes, you can give an offer, but you want it to be small. And then if someone has reached checkout, you don't want to give them an offer at all. You want to do some kind of risk reversal. Maybe you want to give them free

shipping. But if someone's this far, I would be skeptical about just giving away margin when it's probably likely they're going to buy anyway within the next 2 to 3 days. Now, upsells, cross sales, where can you actually put them? You can put them in Cart, you can put them at checkout, or you can put them post purchase. Now, you want to be thinking through number one, the friction of where you're putting the cross cells and upsells. Here, the friction is medium because the buyer is still deciding. So, if you're still deciding whether you want to

buy and then I'm also throwing upsells and cross cells at you, it could potentially make you less likely to buy. Number one. Number two is it's definitely going to make you less likely to actually take the offer cuz you're not bored in yet. So, the typical take rate here is between 5 to 15%. Now, once you get to checkout, friction actually increases further, which is counterintuitive. And I used to think about this the other way around, but our opinion has changed. Take rates decline on a checkout offer. And the reason why take rates declined is

because they already inherently committed to the order and the price that they have made. And so then when you try to force a another offer onto them while they're at checkout trying to purchase, it ends up getting taken less. They're still almost in a browsing mindset here, willing to increase average order value. But at this point they are so committed that you trying to just layer on and stack more things into their cart often doesn't go down as well. Now what actually helps here a lot and how you can increase take rates and do a

good job at checkout is that you don't try to sell them another product. You try to sell them some kind of low friction add-on. One thing that I used to crush on back in the day with this add-on is adding in shipping protection. And so you add shipping protection as a potential add-on at checkout. It's an additional $4.99. You'll get like a 30% take rate on shipping protection and this is 100% margin. You almost will never actually refund on shipping and if you do normally the carrier will pay for it anyway. And so this is

a really easy way to add $1 to $2 of contribution margin or gross profit to all orders with no overheads. On incart you want to be adding usually threshold unlocks are really common. So you have the bar at the top and it's like add one more product and you'll get a particular price. That's where you would place those offer mechanics or you add samples or you add like a complimentary product and this is if you hit the threshold. You then have post purchase. Now there's pretty much zero friction here because they've already made the order.

Payment's already been captured and if it's set up correctly this is a one-click ad. So it's do you want this? Here's the offer. And if you click add, instantly charges to their card and gets added. So this is super low friction. I would always have this. Take rates on this as well can be unbelievably high. I actually know people running 35 40% take rates on post-purchase offers which is so crazy. The other crazy thing is you can string post-purchase offers together which as a consumer gets really annoying. I hate when people do this to me

but as a advertiser I love it because you can give someone a post-purchase offer which is this additional product short time next 60 minutes discount of 30%. you get it taken, then you hit them with another offer and you go because you took this one, we have another one for you which is 60% off and it's another one and you have another take rate percentage on the take rate. And so You might have 25% take rates on the first and then 10% take rates on the second. And once again, I've seen some uh supplemented CPG

brands that have crazy take rates post purchase. It's allowed them to increase average order value by 50 60% which just makes them way more aggressive on acquisition and it allows them to spend double the amount on Facebook. What you can sell here is literally anything. This is just where some good AB testing goes far. I just realized we've gone through this entire segment, but I haven't even defined what the difference is between a cross-ell and an upsell. A cross-ell is a different complimentary product alongside what you're already buying. So, an example is that if you

buy a t-shirt, we go and offer matching socks or matching shorts. This is best for multisq brands because you're selling into other products. But an upsell is just selling more of the same. So, if you're buying the one month supply, we're going to offer you a 3mon supply with some kind of discount. Now, an additional metric that you can track here for post-purchase offers is what's called RPV, which is revenue per visit. You take post purchase revenue captured and you divide by post purchase funnel visits. You can also calculate this by average upsell value times

by conversion rate. Now, the idea here is this is just telling you how many dollars extra are you making out of every person that goes into the funnel as a function of the post-purchase upsell. Now, a good benchmark that you want to aim for is 10 to 15%. If you don't currently have post-purchase upsells, well, the crazy revelation if you made it this far in the video is that you can add 10 to 15% to your average order value overnight just by adding post-purchase upsells if they are done well. It's obviously the caveat of the

if it's implemented Correctly. So for context, if you have a $200 average order value with an RPV of 10%. Then your RPV in dollar value is going to be $20. So you can make an extra $20 on every order through a good post-purchase offer. Now, how should you string a post a post-purchase offer together? You want to always anchor high and then downell below it. And most brands end up inverting this where they lead with the cheapest offer because they think it'll convert better. But what actually happens is that they price anchor at $20 and

then every highv value upsell that they try to do after that looks super expensive. So you want to end up reversing the sequence. You show an upsell of $50 to $60 post purchase. You anchor there. If they don't take it, you downell them into 20 to 30. If they do take it, you bring them over to another upsell which is slightly lower at 30 to 40. And if they take that, you can flow them again into downell number two. or if they don't take this, you flow them from downell one to downell two. The five

rules of upselling in cart is number one, you only want to be having one offer per slot in the drawout cart. Or else, if you have three offers, you just cause decision fatigue, which is actually going to decrease your take rates more than it will help it. Number two is you want to focus on relevancy over price, which is that you can pull in a good price offer any day of the week, but if it's completely not contextual to what the person actually has in their cart, then it doesn't matter at all. they're not going

to take it. Number three is you want to focus on high margin SKs. You don't really want to be upselling into something that's only going to get you an incremental $45$5 in gross profit when instead you could have something in here that's Relevant, that's a good offer that also has high margin. Number four is you want an easy to understand offer, particularly on cross sales and upsells. Unfortunately, you can get as complex as you want in the front-end offer. And that's what the majority of this video has been about is that you actually want to

increase the complexity of the offer because there's way more stuff that you can do than just straight discounting. But when it comes to cross sales and upsells, unfortunately, you don't want to get too complicated because you've got really two to three seconds where the offer is legible. They're going to read it and then they're going to move on. And so you want to make it abundantly clear what the offer is, which usually unfortunately has to be in the form of some kind of flat discount when it's added to cart. And then number five is you

want to measure the success of upsells in cart not by average order value lift but by gross margin lift. Or else you will end up over prioritizing into low margin SKs rather than high margin SKs that might actually have a lower average order value. So moving into how you select the right offer for you. There's four questions you want to ask. Number one is what is the gross margin flaw? You want to start designing offers and understanding what the economics of them actually looks like and whether they will improve the profitability of the business or

whether you're just going to erode margin for no net benefit. Number two is what does the average order value distribution look like of your customers? Is there two peaks? Should you be splitting the offer across those thresholds then to move those customers up or is there just one peak and you should focus on moving one core audience up into spending more like in supplements where you might just have a 30-day supply and you're trying to move them up to 90. Number three is do you have a single hero skew or do you have multiple SKs

cuz this completely Changes the dynamics of how you put the offer together. In single SKS you want to be leaning on product repositioning as the highest lever then gift with purchase and being disciplined with discounting. But on multis you can do a lot more. You can go into native bundles. You can do cross category upsells as well as bundles and you can start to add in a bunch of threshold mechanics that unlock different products. And then number four is what is the buyer's repeat behavior? As we went through before, you can implement subscription offers into

the front-end offer. You can start to really play around with discounting the presentation of gift with purchase, the time periods in which you try to lock people in and build them on in terms of cycles based on the repeat behavior of the customer. You might also have almost no retention or repeat behavior because of the industry that you're in. And so as a product of that, you really want to focus on actually sustaining gross margin on first order, considering that's where most of your gross margin comes from. One last comment here on selecting the right

offer for you is that there is 100 different offers that you can run, there are probably 10 that are good. And then of the 10 that are good, no one can really say which one will work best until you actually test it in market. Then off the back of that, once you have a large enough existing customer base, you want to survey the existing customers as much as possible to be able to further iterate on the offer. Offers also typically don't stay static over time. So if you look at any large CPG brand, the offer

may stay static for 6 months, 12 months, but on the side, they are likely split testing different offers on dedicated landing pages and funnels to then find a better one to rotate in. Because the unlock in being able to find a better offer is one of the most valuable things that you can do in the front-end funnel as long as creative and everything else is dialed in. So then lastly, there's seven ways That offers end up failing. Number one is people just default to a blank discount and they call it an offer. They said that

our offer going into this period is 25% off, but it completely negates everything that we've gone through in this video, which is that the offer is way more than just a discount that you apply. Number two is people miscalculate gross margin. They either just include cost of goods sold and so they don't include the impact of shipping and fulfillment, which might actually improve in bundles and as units per transaction increase, or they calculate it wrong in the other direction. And so they don't understand the true impact of gross margin on efficiency targets like rorowaz and

therefore they discount way too hard to where the efficiency will never be able to reach the discount threshold to be able to maintain margin. Number three is that people run the same offer to new customers and returning. Fundamentally, these are two very different mechanics. On new customers, we're optimizing for first purchase economics and getting the customer in to buy. on repeat offers. We're really trying to deepen cohorts by locking these people in to buy for a sustainable period of time. And so the offer that hits these two audiences should look very different. Number four is

people set their thresholds based on their mean average order value rather than looking at the distribution curve. Honestly, probably like 80% or more of people do this and it means that their offer is just completely set up wrong. Number five is gift with purchase is actually set up with the wrong gift. either it's a gift that their customers actually don't care about and it's not materially even changing take rates on this offer or this offer just isn't high margin enough that it's not actually Changing the economics of the bundle substantially enough to even matter and

so the economics and the math wasn't ranked correctly behind this gift. It was just thrown in because gift with purchase sounds good but we didn't actually consider the economics of it. Number six is just running a 25week promo calendar all year round. continuing to just rotate out different percentage discounts across different product categories and calling that offer rotation when this is incredibly unsustainable and means that you just have to do this forever. And then number seven is treating Q1 as if it's a seasonal dip and not an offer issue. You can actually rotate offers into

Q1 that are contextually relevant to this point in the year that allow you to not have a dip. It's also one of the best points in the year to be testing offers because of the overall suppression in conversion rates and market sentiment. So the five things I'd be taking away from this video is number one, the offer becomes one of the most important levers in acquisition. And so if you're not testing your offer, you should be you should be setting up separate funnels with dedicated landing pages to test different front-end acquisition offers as it can

substantially change the profit on first purchase as well as what retention looks like. Number two is the offer isn't the discount. The discount is one mechanic that sits inside the offer, but it's usually the worst one to lead with. Number three is that discounting without understanding the curve of how contribution profit and average order value changes means that you won't plan correctly and you'll just set up discounts. You'll set up offers without understanding the other side of the coin which is what efficiency do we have to hit on paid platforms to be able to actually

make this work. Number four Is the five offer mechanics which is product repositioning bundles gift with purchase tier threshold and then disciplined straight discounting. And then number five is you should have some form of upsell, cross-ell or post-purchase upsell. And it should be structured around maximizing contribution margin. And I gave you two metrics as to how you can actually measure success. Now, if you want a calculator that shows you how your discounting changes your efficiency targets, we'll put a link in the description below. If you're an e-commerce brand, you can access it to start planning

out your offers and how discounting is going to impact efficiency. If you're an e-commerce brand doing over $5 million a year, you can also click the link in the description, watch a quick two to three minute video that'll run you through the audit process. And then if you're a performance marketer that has gotten this far in the video, reach out to us at hiringdigital.com.au. We're always looking for new performance marketers on the team. If you run meta ads as an e-commerce brand, this is the only video that you need to watch on meta structure. Whether

you're just starting a new ad account or you're spending a million dollars a month on Meta, we'll be running through every level of structure, everything you need to know so that you can maximize efficiency on the platform. There's three claims on account structure that are all true at the same time. And by the end of this video, you'll understand why three of these claims all make sense contextual to your ad account. Claim number one, consolidation will always beat segmentation within any meta ad account structure. So, if you have fewer campaigns, fewer adsets, you'll generally get

better performance, but it's not necessarily the right thing to do. Number two is that media buying actually still matters at higher ad spend levels. Now, if you're spending $100 a day, it's probably not that big of a lever when it comes to growing the Account and growing the business. But if you're spending $100,000, $200,000, $300,000 a month, maybe you're buying is still going to give you a 10 to 20% efficiency lift if you do it correctly. And then third is there is actually no universal perfect account structure that fits every business. The reason why there

is a thousand YouTube videos on how to structure ad campaigns is because there is a thousand different types of businesses. And so dependent on the nuances within your particular business, it will change how you think through structure. I'm going to be giving you through the way that you should think through the problem so that you yourself can make the structure yourself. Over the next 2 hours, I'll walk you through how Meta decides where your money's going. The three spend playbooks if you're doing under 50k a month in spend 50 to 250 and 250 plus. The

difference between ABOS and CBOS and when you should use each. Three settings that you should be turning off right away as you finish this video. How to design high performing adsets. How much spend should go towards existing customers and different audiences. What's the testing budget look like? How do you math that out? How do you back propagate from your goals? And then number four is I'm going to give you four diagnostic questions that you should always be asking yourself. Meta is optimizing for one thing in 2026 and honestly forever. It's revenue per user per minute.

Now there's two ways that Meta can grow this number. Number one is they increase ad inventory. And so they simply put more ads onto their platforms or they buy more platforms where more ads can serve. Now you've actually seen this be the case over the course of the last six years. I remember six years ago I would get one ad in every five to six posts. Now, sometimes you get triple ads. You're scrolling and you'll get an ad, another ad, and then another ad all in a row. And that's meta increasing the ad inventory so

that they can maximize revenue on the platform. And then the second is if they can't increase ad inventory any further, if the platform's starting to get not enjoyable to use because there's just so many ads. Well, the other thing that they can do is they can just increase the cost for the advertisers to serve, which is your CPMs, your cost per thousand impressions. And so the same amount of ad inventory, but let's make it more expensive for everyone. And you see this in year-to-year CPM inflation. We have over 150 million in ad spend connected to

our business manager. And we can go and do an aggregated ad report and look at what CPMs have looked like over the course of the last 3 years. And it's cyclical with obviously Black Friday, but every year it goes up. And what you end up seeing is there's about 20 to 30% inflation in CPMs. Now, obviously, there's a little bit of natural inflation to the dollar that you need to factor out of that, but still CPMs are going up faster than inflation is. And the reason for that is that Meta has to continue to publish

good quarterly earnings because they're a publicly traded company. And the way that they do that is they need to increase revenue on their biggest product, which is the ad product. Now, this seems all very doom and gloom and like, "Oh, Meta is against you. Everything's becoming more expensive. It's a terrible platform. Don't spend on it." That's not necessarily true because if Meta increases CPMs, everyone just becomes unprofitable. So they can't do that. They can't just make all their advertisers unprofitable or else people will stop spending. And so what they have to do when they increase

CPM is they also have to increase expected conversion rates or ROI of the platform. So they need to effectively more efficiently pull dollars out of users on the platform and transfer them to advertisers so that you get conversion rates that outweigh the CPM increase. The reason why all this matters and the reason why this context matters is that every time Meta rolls out an update, whether it's Andromedor, whether it's Gemin whether it's one of the smaller updates that you've never even heard of, what they are trying to always do is increase the efficiency of ad

serving so that then they can make it more expensive for you to actually place. Now, a core principle of setting up structure on the account is understanding that what you see in the platforms is not necessarily what Meta is optimizing. And that's where you get a lot of poor decision-m being made within account structures or within decisions as to where budget should flow. There's a lot of stuff that you see that's not necessarily reality. And so an example is what you see in the platform is lastclick attribution. Meaning if a user clicks on multiple ads

and then buys the purchase the conversion value will just go to the final ad that they touched. But in reality Meta is optimizing across multi-touch attribution. Meta knows that just optimizing towards the last click is meaningless when there was all of these prior clicks and prior interactions that led up to the conversion. And so what you end up seeing is that if you have three ads here and the clickfunnel looks like this and then they ultimately buy, you on the surface will go, "Ah, this is the ad that's performing killer." But Meta will still be

distributing spend here because it sees that a click actually occurs here, click occurs here, and then the final click occurs. And so in the back End, it's likely allocating 1/3 credit to each one, which is why spend's getting distributed in the way that it is. So another example of what you see is that one ad gets all the credit. But the reality is that Met is optimizing across a chain of impressions. What you see is ad level return on ad spend. So you're going down here and looking at rows on these individual ads. But what

Met is actually optimizing for is your CPA target at the adset level. Because the bidding and optimization actually is inherited from the adset, which is why you don't set cost caps at an ad level. If you're going to set cost caps or any kind of more complex bidding approach, you're going to do it at the adset level instead. Now, if we go back to this concept of sequencing where a user might pathway through multiple ads, all of the ads, let's say, are spending $5,000, but they're at very different returns. When you look at this on

the surface, you make an obvious decision if these are all sitting under the one adset, which is let's turn this ad this ad off. Let's put all the spend here. Now, besides sequencing, besides the fact that, well, these ads are probably doing a little bit of heavy lifting, we need to think about why is Meta distributing spend to these two ads. Well, there's something called the breakdown effect. And this is public documentation by Meta. You can go and Google for it right now and pull up their actual uh page on this. And the idea is

that when you use any of the breakdown features in Meta Ads, so you can go to breakdown and you can click on age or you can click on gender or you can click on placement type or you can click on region. Okay, there's all these different options as to how you can segment and break down the data. What you will often see is weird stuff that doesn't make sense. You'll look at a breakdown, for example, on placements, And you'll see that stories are at a 4x row. Feed is at a 3x, but the feed is

getting all of the spend. It might be holding 80% of all the spend in the account. And you go, why is this happening? Seems pretty easy to fix this, right? We just launch a dedicated campaign that only places on stories because stories perform better. But it's a flawed assumption because you are looking at blended data that is not counting in the incremental impact of pushing more spend through that particular channel or through that particular placement. And so if we simplify this back to the ad example up here, the reason why this ad is getting the

same amount of spend as this ad is because Meta has tried to spend more here. It's tried to push it up past, let's say, $300 a day to 320. But when it does that, there's zero incremental returns. You don't make any more money. But over here, when this was, let's say, a lot lower down at $200 a day, Meta went, "Oh, we can't put spend here. We're getting no returns." So, let's see. Even though this has a lower base return, do we get any incremental returns here by putting more spend? And it goes on layers

250 in and we make more revenue. Now, it's not at a great return. Maybe it's at a 2x or something, but this is at a 0x. And so, budget goes here. Same thing applies at a story and a feed level. So Meta will try to put more budget through stories. Obviously, it's got a better return, but when it does it, the incremental ROI is so poor that it would rather just put spend into the feed. So the distribution of spend Majority of the time at a breakdown level is actually accurate and you should trust Meta.

Now, there are cases where that isn't the case, and that's where you have to be a good media buyer and understand what breakdowns matter, what don't, and how you should be thinking through distributing spend. But majority of the time, if you're a beginner and you get too deep into breakdowns, you'll just make a bunch of segmentation decisions that actually aren't commercially aligned with what the platform wants you to do, and you'll get worse performance. So unless you are like an expert six year, sevenyear in media buyer, you should not be implementing changes based on complex

breakdowns or your limited understanding of return on ads spended an ad. Now it wouldn't be a Facebook ads long form video if there wasn't a mention of the buzzword Andromeda. So here's a 30-cond explanation which is that previously you would choose interest, you would choose audiences and then you would load up a bunch of creative and that creative would serve to that interest. Now you load in your creative, you leave everything broad and Meta looks at the creative and based on its understanding serves it to a relevant audience. That's effectively the retrieval system chain. Now

you might be thinking, well, does that mean interests are dead? Does that mean lookike audiences are dead? We shouldn't use these anymore. Generally speaking, yes, interest targeting is super flawed. And if you actually go down this rabbit hole, you'll learn a bunch of reasons as to why interest targeting is not good. I'll give you one of them, which is that interests don't take into consideration intent. And so if I said I hate dogs on a Facebook post, Meta would group me into being interested in dogs. And so if you go and target the dogs interest,

you would target me. But I actively went out there and said, "I hate dogs." And I commented on a bunch of posts. And so it isn't Intent driven. It's simply interest driven. And so because of that, Meta put out a report or someone put out a report saying that 30% of the people that are inside of an interest group aren't meant to be there. They were incorrectly assigned. And so interest groups in itself is a poor way of categorizing users. Now, if you really want to go a little bit deeper on this topic, the way

that interest targeting worked conceptually is it was effectively labeling. And so you would be interested in a pets post. And once again, that just means an interaction, comment, a like, something. Meta would tag you with pets. And now anyone that wants to target pets would target you. Very rudimentary. When you think about Meta being a trillion dollar business, you're like, "What? They're just putting labels on people based on the post that they interact with. That doesn't seem very sophisticated and it's not, which is why it doesn't perform as well as targeting broad. What happens when

you target broad? Instead, you can think of it as if it's a vector space. Now, I'm just drawing three axes here, but in reality, this is like 10,000 dimensions. But what happens is that when I go and interact with a pet's post, I get plotted on the graph. And then depending on what I interact with in real time, I will get moved around on this chart in a certain direction. So if I interact with cats as an example, I might get moved up in this direction. But if I interact with dog posts, I get moved

over in this direction. And then what ends up happening is people will naturally cluster through this three-dimensional space. And then when you go and target broad, instead of targeting people with labels, you're just targeting this big space and like sprinkling your ad out all over the place and going who interacts with it. And there'll be these hotspots and over here the people in This area actually start buying from you. and Meta goes, "Okay, well, let's just target this area." And then it will go and target this area right here. And then as you start to

scale, what happens is the area increases around this spot. And so you target colder and colder audiences. And this is why most ads fatigue. This is why most campaigns die when you try to scale them up because you go from a hyperspecific target demographic and you try to scale out of it. And these people out here actually aren't convinced enough to buy your product cuz your ads might not be good enough. Now, the caveat with this whole model is that it isn't three-dimensional. it's like 10,000 dimensions and so it can be much more specific in

the actual clustering of users. So the first core thesis to understand is that consolidation beats segmentation. Really important concept to always be thinking about anytime you're structuring a meta account. Now I want to run you through the timeline of why this used to not be the case and why it's counterintuitive and why a lot of agencies particularly legacy agencies are still hyper segmenting all over the place and it's probably ruining your performance. Now, what's also really unfortunate as a little side note is that the legacy outdated agencies are typically the very cheap agencies, which means

they're the agencies that work with smaller businesses. And so, as a product of that, a lot of smaller businesses that might be watching this video that are just starting or maybe spending $10,000 or $20,000 a month on Meta, if you're with an agency and they're cheap, they're probably running legacy structures. And so, this is probably relevant for you so that you can push towards what actually works. these days. So 2018 was actually the first time I opened up a Metat account and I was spending my own money at the time. Now back here you could

segment by interest. In fact, it was favorable. Interesting was actually one of the biggest levers in the account back here. Now obviously creative still mattered. Obviously the website still like all of these things played a part. But there was an additional lever which is that if you could take a creative and you could find the interest that it performed on, you could achieve scale. and creatives would not perform on some interests and they would perform on others. In fact, you could duplicate adsets with the same interest, same ads, and sometimes they'll work, sometimes they won't.

And so, there was all these different nuances in the account that would actually allow you to see performance. And the reason being is that the platform was built around this. How it used to work is every time you would launch a new adset, the adset would go out and it would serve to a thousand random people and based on the initial intent signals, so who would click, who would interact, potentially who would buy, meta will then zone in on those types of users within the interest group. And so you might have had that when you

launch this adset, for whatever reason, moms age 40 to 50 interact. And so it starts going after that audience. You could launch the exact same ad set a second time, same ads, same interest. But on these thousand people, for whatever reason, men aged 30 to 35 are the ones that interacted because it's a very small sample size. So you can have bias in these small sample sizes out of the gates. As a product, this adset starts going off and optimizing towards those types of audiences and you end up with very different results on each and

you end up with very different audiences, too. And that was just a product of the way that the ad serving worked at the time. And so you were Favored to Seg. It was a good thing to have a ton of adsets where you were constantly doing testing across different interests, across different audiences, across lookike audiences so that you could try to squeeze out more. Then what ended up happening? Well, Meta improved the ad product, right? We're looking at an 8-year time horizon here. Some stuff happens. iOS 14 popped up which substantially impacted the ability for

Meta to actually do interest groups because of a lot of the interest grouping was done using offsite pixel data from blogs from whatever they were interacting with on the internet. So they had to start to change the way that their retrieval system was working as well as their ranking system for users. Then AI started to pop up over here and that's when Meta became meta. Went from Facebook to Meta and Zuckerberg went and made that big play into the metaverse. bought a ton of chips off Nvidia. They ended up not being able to do anything

with those chips. So what did they do? They rolled them into inference to be able to better train the actual ad surfing platform. And so then the ad serving platform started to improve even further and they started to be able to squeeze more efficiency. But then as a product of that consolidation was preferred during iOS 14.2. Just a side note, Meta lost they said about 30% of the signal that they were using for targeting at the time, which means that they had to compensate for this by finding better buyers by not relying on interest targeting

signals and instead being able to use the creative and go broad. Now, the fundamental reason why The platform now prefers consolidation is because conversion data is siloed at the campaign level. This doesn't just apply to Facebook, it applies to Google, it applies to Tik Tok, it applies to Pinterest, it applies to all of the ad platforms. They're all structured in the same way if you haven't noticed which is you have campaigns, you have adsets and you have ads. On Google you have campaigns, you have ad groups and then you have app. This is the same

case on all the platforms. And the reason being is it's the way that the platform silos out data in targeting. I won't go too deep into it, but I think it's important to understand at least the architecture of an ad account, whether you're a beginner or whether you're an expert because most people never even think through this problem. Um to be able to better understand structure. So you should always be challenging everything. Why does this exist? Why are they doing it in this? Why do campaigns and adsets exist? Why are there adsets? Why aren't they

just campaign? In fact, why isn't there just one camp? Why isn't the platform just ads and you just input your ads and there's no structure up here? Like, why does all of this structure exist? Why have they introduced this additional complexity? And it's because you do need a level of segmentation within the account to better commercially align with the objectives of most businesses. The reason why campaign segmentation exists, the reason why you can make multiple campaigns is because businesses often need to flow budgets through different campaigns over time. And so you might have a promotion

that rolls in that then rolls into something else. And so you Need segmentation in the actual KPIing and reporting. Number one. Number two is different business units exist in a lot of businesses, which is that you might have a category for pets and then you have a category for children. As a product of that, you don't want the campaign getting confused and not understanding who to target when you have slightly different demographics. And so, as a product of that, you want the ability to be able to segment the account based on the actual segmentation of

personas within the business. And this allows you to do it. And one of the downsides of that is that conversion data here is not shared at the same level as it is at the adset and ad level. And so is there some conversion data sharing going on here? For sure. Campaign one is pulling from some of the signals of campaign 2 and vice versa. But they're not directly sharing all of the conversion data with each other. Which means that if you start to introduce a bunch of campaigns in the account and you have let's say

four campaigns for a small ad account with not much budget, what you're actually doing is taking all your conversion data. Let's say you're getting 100 conversions a month and you're splitting it up across four campaigns. So now you have 25 25. You're inherently going to get worse results in the account due to segmentation because now you have less conversion data that these campaigns can optimize on. And anytime you have less conversion data, you will see worse results due to small sample size bite. Which simply means that if you don't have a lot of conversions, meta

doesn't have a lot of signal and so it doesn't really know who to touch. The more signal you have, the more conversions you have, generally actually the better the efficiency is. Which is why you often actually see that when you can crack through 100 or 200 conversions a month, the ad account starts performing better rather than worse. And it's because there's enough signal that Meta now understands who to target properly. And you can get out of that rut of Meta not really understanding who to target because there's not enough conversion data. So you always generally

want to be consolidating up at the campaign level cuz every time you segment, you're chunking out the data. Same thing kind of applies at the adset level, not as much. There is more data sharing that occurs here than at the campaign level. Number one. Number two is that all the data at the campaign level also inherits down. So you don't really have to worry as much about segmentation at the adset level. You can kind of segment all you want. It's not going to impact performance that heavily. You can consolidate up. It's not going to perform

impact performance that heavily. We're talking about really 5 to 10% efficiency swings here based on hyper segmentation versus consolidation. It's noticeable particularly at scale, but it's also going to be context dependent on the actual. Then at the ad level, obviously you're going to have massive segmentation because you can't stack ad. Now you technically can stack ads and the name of this always changes. So by the time you're watching this, it might be different. At the moment, it's called flexible ads, which is where you can load in like five different creative on the same ad. The

reason why we generally don't like these is because you don't get visibility into data insights. And So you can go and load five creative up into one ad, which is all well and good. Nice. It's consolidated. Meta prefers consolidation. So it seems like the right idea. But the issue is when this ad performs well, we don't know what creative is actually performing well. Is it creative one, two, three, four, five? which of the creative that we put in here is actually lifting performance because we want to do more like that. We want to make more

ads like that. But if we don't have the ability to read the insight, well, it's kind of useless. Yeah, we got performance, but now we don't know what to do with it. There's also a couple other benefits of consolidation over segmentation. And this is mainly letting the machine decide where to distribute spend. Now, this can be a bad thing and it can overweight your account into a heavy degree of risk, but can be a good thing if you're just trying to maximize efficiency in the short term, which is that when we're talking about consolidation, Meta

is picking where to distribute your spend. If you just have everything sitting in a campaign that's a CVO with a bunch of adsets with a bunch of ads, Meta is just going to go and swing budgets around based on what it believes is most efficient. when you have segmentation and let's say an structure. So you're choosing where budgets go at the adset level. Well, you are deciding where the budget goes and inherently meta will always in most cases distribute budget more efficiently than you will. However, Meta will also therefore distribute budget on an 80/20 Purto

sprint, which is 20% of your ads will get 80% of the spend. Now, kind of annoying if you're making a 100 new ads a month and none of them are getting any spend and they're not getting tested. Also kind of annoying if all the spend goes into just one creative and maybe you're paying influencers for partnership ads that are getting no spend or maybe you know that creative is Going to fatigue and now you're like we're in a risky position because we have no backup ads that are doing well. And so this can be good

but it can also hurt you. Whereas on segmentation you're going to be forcing spend across a structure in a way that you want which technically might derisk you might put you in a better position might facilitate better testing. uh if you're a novice and you're not doing this thoughtfully and methodically and carefully, well then you can just end up burning a bunch of money here. And so a good way to kind of think through this problem is that if let's say you're an in-house business, $10 million a year and you're making an in-house media buying

hire. If you're hiring someone that is in their first year and they have kind of no idea what they're doing, I would prefer they run this structure cuz I would actually prefer Meta Distributes budgets over them because they don't know what they're doing. If you know what you're doing and you're experienced and you understand weighing risk against profit and efficiency and understanding the changes that need to be made, I would much prefer they run this structure or somewhere in the middle so that we can still have control over where spend's getting forced carefully and thoughtfully,

but we're not just burning money. There's also the extreme of this which is there is a circle on the internet that is focused around consolidate everything. Just run one campaign, one ad set, throw all the ads under it. That's all you need. You don't need an agency. You don't need any kind of media buying. You don't need any kind of segmentation. You don't need structure. Just consolidate. Just upload ads into a campaign and you're good. That's a terrible idea for 90% of people. Now, 10% of brands can get away with that and they'll be okay.

But there's three reasons as to why you don't want to just consolidate everything. Number one is different unit economics. And so, if you're a large business with a lot of SKs, generally The unit economics or the gross margin will change across different categories. you will have 70% gross margin on this particular product range. But on this particular product range, maybe you have 60%. Or you will have grade A, grade B, grade C, grade D inventory. So on your grade A inventory, you have incredible unit. On your grade C inventory, you also have incredible unit. Your

gross margin is the same. It's great, but it's not selling. And so because of that, you generally have to discount it hard, which compresses the actual margin post discount. And so if we put both of those products or both of those ads into the one campaign, we would end up with likely a lot of spend getting distributed to the heavy sale items. But the heavy sale items is not where we want our budget to go because we have compressed margin there. We're probably going to end up with worse contribution margin and it's also just bad

branding to be heavy pushing sales messaging through our entire funnel and existing customers. So when you have different unit economics across the product portfolio, you actually want segmentation. Number two is you might have different audiences. And so an easy example of this means that you might have an activewear brand and the activewear brand sells to both men and women. Now if you go and put all the men and all the women ads into one adset into one campaign, the adset is likely going to get confused. And in fact, you see this on ad accounts all

the time. This is an easy call out anytime I audit a brand that sells to both men and women through core different product ranges. So there's a men's range and there's a women's range is that you just go down under the adset level and you just take a look at all the ads. You use the breakdown feature. you look at gender Split and what you'll end up seeing is that women ads are going to men and men ads are going to women. Now men ads serving to women isn't as bad of a idea because women

do a lot of purchasing for their partners and so you end up seeing directional purchase behavior from women to men but you don't see it as much in the other direction dependent on the category and so you generally don't want to be pushing women's active wear ads to men. Will you get a return? Sure. People will buy. You still get a return on it. Is it the most efficient use of your capital in the business? Absolutely not. You should serve your women's active wear ads to women. And so if you just go and put all

of these in the one ad set because targeting primarily exists at the adset level, that is where you choose the targeting. That is where the targeting generally resides for most of what's occurring under it. It gets confused. It doesn't know where to serve them. So, it's just serving them to everyone. Number three is high average order value or low conversion volume accounts. We have some clients that we work with where average order value is $10,000. on a $10,000 average order value. The issue is is that you just don't do much order volume. And so this

brand as an example, I think they do 50k a day or something like that, it's only five orders a day. Not a lot of purchase volume at all. And so what ends up happening in an account like this is that if you have a CBO as an example, and this is a bunch of ad sets, so this is ad set one, set two, etc., is that Meta will distribute budget at the adset level not based on purchase, not based on row, not based on actual signal that we care about. Instead, because there's nowhere near enough

signal, there's only like five conversions max getting attributed into the account per day. What ends up happening is Meta needs to go upstream in its optimization. And so, it goes up and starts looking at CTRs and CPCs and hold rates. And now sure these are pre-intent pre-click signals as to something that might infer future performance. But unfortunately the reality is when you look at very large data sets CPCs are not correlated with returns except for the extremity of bounds. And so if CPCs are super high yes ROI will be super low. If CPCs are super

high that actually doesn't correlate with ROI. So what you end up saying when you graph this out is you just say something like this, which is that cost per click has almost no correlation or a very weak correlation to actual return on the ad. And so if we're optimizing budget distribution based on cost per click or CTR, we're optimizing it towards a metric that has nothing really to do with the actual core objective, which is to drive revenue. And so in this case, we likely don't want to use a CBO. We likely don't want the

campaign to distribute budgets how it is. And we want to think through a structure that's going to look very different from what you would expect in any other account which is sort of getting us towards the position of every account looks different depending on the actual commercial objectives of the business. What do the unit economics look like? What do the audience profiles and personas look like? What does the average order value and conversion volume look like? Because all of these things will change how we approach consolidation or segmentation. I'll give you a tactical example of

where consolidation on an audit I did a few weeks ago was very relevant and not a good idea which was this was a furniture brand about 20 to 30 million they had a cold campaign which was consolidated that's all they had and then it was a CBO and they had a bunch of adsets down here now the issue was 57% of the meta budget was going into the top adset now what was this adset it was just a dynamic product app now the issue is return on ad spend looks really good on service it was

like a seven rorowaz but once you break down to 7-day click or you break down to incremental attribution either one gave the same read returns was actually a 1.9 everything else was performing better than this on an incremental or a 7-day click rate and so this was a good example of where them just consolidating up and putting DPAs in with everything else not a good idea particularly in that industry because you'll always end up with overspend into an ad type that doesn't actually perform that well on cold audiences let's talk about what in media buying

actually died what doesn't matter anymore and then what still matters because there's a lot of stuff that is irrelevant these days. And when I say media buying still matters, I'm talking about a very specific subset of things within the account. There's a lot of stuff that you can push to the side. What 99% of people shouldn't be doing is daily bid tweaks. Going into the account using cost caps or bid caps and just changing bids every single day. Now, that used to be a strategy four or five years ago that a lot of people did

quite well with, but these days you're really moving the needle on a variable, on a lever that has nowhere near as much leverage as other things that you could be focusing on. Now, could you technically squeeze 5% more out of an account using daily bid adjustments every single day and dropping an hour a day into this? Sure, probably. Could you spend the same amount of time just making better creative and double the account or triple the account? Definitely. And so if we look at return on capital being time allocation, you're way better off just making

better, more creative than going in there and just Tweaking with bids all day. Now, what still matters is account structure and account architecture. So thoughtfully thinking through how we're segmenting and setting up the account in alignment with the commercial objectives of the business. The next one, for smaller accounts, this is applicable on very large accounts, but for smaller accounts, daily budget pacing. So changing budgets every single day in alignment with let's say expected conversion rates across the week or just in alignment with daily performance, not a good idea. You don't want to be going in

and tweaking budgets all the time. You're going to mess around with campaigns. You're probably making these decisions on very limited sample sizes of data. Once again, was this the strategy that you would do back in 2019, 2020? Yeah, for sure. Is this something that matters these days? Not at all. What still matters? Concept level segmentation at the adset level. We're going to go into this in a lot more detail later on in this video, but this matters a lot. What doesn't matter on the other hand is interest targeting. Very strongly of the belief this shouldn't

be in your account. You're wasting resources. you're wasting tests on something that doesn't even really work anymore. What's a better use of time is instead of going in and doing interest testing and interest targeting is you can do page testing instead. So split testing different landing pages, split testing CRO on the website. This matters a lot. You can make an inflection in conversion rates. That's a much better test than going and just messing with interest targeting on the platform. I'm also going to throw in here lookike audiences don't matter anymore. They don't work. I could

once again talk for 5 minutes about all the deficits of lookalike audiences and why you should just be going broad instead. What does matter on the other hand is thinking through and controlling percentage or dollar allocation to existing customers. This is something That you should really be thinking about and controlling over time. Before I start to give you the actual account structure that you should be using at each level of spend, I want to quickly talk about page strategies. We as an agency only work with eight and nine figure businesses. And so as a product

of that, the content is more tailored to very large businesses and people that are in a position to work with us. If you're small, this is going to be kind of irrelevant. If you're big, this is going to be super relevant, which is that Meta caps the amount of ads that you can have live on a page at about 3 to 500 ad. And so what this means is that if you launch a lot of creative every single week, every single month, you very quickly pass this cap. If you're even a somewhat big account, you

then have two options. Option number one is you just start turning ads off because you can't have more than 500 active ads in the account which is really annoying. Or option two is you make more pages. And so I'm going to give you three options here that we recommend to all clients when we run into this position. And we run into this position with pretty much like every single client because all of our clients are launching enormous amounts of volume. Number one is you create multiple duplicate pages. So someone that does this quite well and

I've done a reel on this on Instagram is Grooms. They have like 15 grooms pages and the only difference is that the logo is a different color and then they're just running ads to a bunch of these different pages. Now, personally, if I was running an e-commerce brand, I would just do this. I would just spin up a bunch of pages for the sake of ads. I'm not really too concerned about consolidation onto a single page, but I know some people are in which case your other option is whitelisting or partnership ads. And so, you

just have to lean more into running ads through other people's handles that aren't yours. Now, worth noting there is a core difference between whitelisting and partnership ads. Partnership ads is where you're using an influencer or a Creator or someone that actually exists out there and you're running the ads through their page. Often you'll have to pay them to run the ads through their page if they have somewhat of a following and they have leverage. Whitelisting on the other hand is you are taking a third-party page that you can create and you're running ads to it.

So in this case, rather than Grunes making another page called Grunes, instead they might make a page called fiber health magazine or something like that and then they're running the same ads. I personally would change them a little bit to be more advvertorial and orientated around this third party framing, but you could run the same ads and instead you're running it through the handle fiber health. And now the obvious advantage there is that it looks like it's coming from a third party. It looks less like you're being sold to. It's a really good strategy that

we use across quite a lot of clients and it's a way that we can decrease the meta ad cap on the primary page. And then lastly, you can also do regional pages. Now, if I was running a brand, I would do all three of these. I would have whitelisting. I would have as many partnership ads as I can. I would be having multiple different primary pages and I would be going into regional pages, which is that when you sell in multiple countries, you hit the meta ad cap way faster. The reason being is that generally

you should segment your campaigns based on country. And so you will have a UK campaign, you will have an Australia campaign, you will have a USA campaign. If each of these has a 100 ads in each, you hit the cap. Even if they're the same ad, you hit the cap, which is really annoying. And so the way that you fix this is you have a Grun UK, Grooms Australia, Gruns USA, or whatever your brand is. In that way, you have new caps per country. And so you don't run into this issue. So the third core

claim I made at the start of the video was that there is no universal account Structure. So, what are the rules and what are the mental models that you should use to be able to think through building your own account structure? Well, there's two non-negotiables, which is number one, data integrity. Now, what this means is that the structure that you use, more specifically, the structure must be readable and actionable. So, whatever you're running, it needs to produce data that is readable and actionable. Most people create account structures, set up their Meta account, set up their

Google account, Tik Tok account in ways that aren't readable and therefore aren't actionable. And so when we try to make decision loops where we look at the data and then we make a decision and then we make time go by and we make another decision, etc., this isn't a viable option because the account wasn't set up in a way that had data integrity. Sounds really obvious, but you'd be surprised if you don't think through this as a core non-negotiable of the structure. you end up leading yourself into a position where you regret the way that

the account's structured and you have to do a restructure. Then the number two non-negotiable is commercial alignment. The structure needs to actually fit towards what the business sees and thinks about its products margin and goals. If a campaign ends up hiding a product or hiding a category that needs spend or it's going to turn into grade inventory, that needs to be presented within the account structure. If there's a product portfolio with a low margin base that actually can't support paid media acquisition, well, that needs to be reflected within the account structure. If there's a particular

goal across a category that needs rep prioritization or KPIing separately, that also needs to be baked into the Account structure. There's five questions you should be asking yourself when you're building out the account structure. Number one is how many product categories do we have? The more product categories, generally the more segmentation will need to be introduced if there's different goals across those categories. Number two is how wide is the margin spread? Is all the margin relatively the same across the different product categories or does it vary substantially category to category? Number three, do we have

multiple personas that don't overlap? Now, it's fine to have personas that overlap and are kind of the same person, but we're tweaking the audience slightly. But if we're talking about very different personas that don't overlap, that will need to be introduced to some degree into the structure. How many regions are we selling in? As I said before, when you're selling in multiple regions, you generally want campaign segmentation so that you can control spend distribution and sell through rates. If you're holding inventory in particular countries and they have their own revenue targets, well then you need

the ability to be able to distribute spend and the only way to do that is to segment. And then lastly, what's the creative throughput? If you're putting 10 ads in the account per month, there's honestly not much segmentation that can be done. If you're putting 2,000 ads in the account per month, it's a very different volume of creative that's flowing through. So we need to think about how is that creative actually going to be introduced into the structure methodically have spend allocation and then give us the ability to move it into a scaling structure. So

let's go through two actual examples. Two different brands very similar in terms of revenue similar in terms of gross margin but the account structure out of the back of it is going to look very different. So we'll build out what the account structure will look like for brand A and brand B. So brand A is in CPG. So they're selling consumables. There's only one product. They're doing 5 million a year. 70% gross margin which is fairly indicative of CPG. There's only one persona so far, which is actually a good thing. Most people at this revenue

level will try to squeeze a bunch of different personas out, but they're only at five mill a year, so they only really need one persona, and that's Busy Moms. And they sell an AU in US. In my opinion, they should only be selling an AU at this revenue level, but they've decided to expand. Brand B is in fashion, 50 SKUs, same revenue, slightly lower gross margin, which is likely indicative here of them needing to discount constantly. Their gross margin shouldn't be this low in fashion, but it's normally a product of the fact that they're taking

a lot of product to discount. They've got four personas. This is fairly typical in fashion, having three to four core personas, but they're selling an AU NZ and UK. So, how do these end up getting structured differently? On brand A, we likely just want one advantage plus cold campaign. We want to be doing creative testing at the adset level based on concepts for a majority of spend should be going into the one concept and the one persona that's actually performing well at the moment. There should be one of these for the US and then we

should have another one for AU. We want existing customers excluded from both of these cold campaigns particularly in CPG where there's hopefully going to be a lot of returning customer revenue that's going to overattribute into cold and the frequency on these campaigns will end up being driven up and it will just retarget. So we want existing customers excluded. doesn't mean we shouldn't have any spend towards existing customers. There's likely a bit of incrementality here in allocating a little bit of spend to them. And so, we're going to throw in a retargeting campaign on existing Customers,

but it's going to be at a very low spend to the point that we just want to keep frequency at below a seven on a 30-day rolling period. So, this will end up at this size of business being a very small campaign, probably spending $20 a day. So, that's brand A. What about brand B over here? Due to the higher complexity of SKUs, there's probably multiple categories here. In fact, there's three. And so, we're going to build out campaigns for each different category, particularly because they have different purchase orders, different revenue goals, and they need

to sell through all of the categories or else they have to go to sale and a road gross margin. So, they're going to have a pants campaign, toss campaign, dress campaign. Now, for AU NZ, we're just going to consolidate this into one campaign. So, we're going to target Australia and New Zealand together. The nice thing about New Zealand as a pairing country is that it will automatically max out at 7 to 10% of total spend. So, we don't have to worry about it overspending and most New Zealand orders will always get fulfilled from an Australian

warehouse with relatively fine shipping crazes. And so, we don't have the need. Now, this is already a lot of segmentation for a brand of this size. And so, we also need to be thinking through, do we really want the UK in a separate campaign? because that's going to mean we have six campaigns now because we have to duplicate the whole structure. We really have two options here. Either option number one is yes, we do that. But rather than having six campaigns because that introduces way too much segmentation in an account of this size, instead we

move the segmentation down to the adset level. So we have one Australia campaign, one UK campaign, but Then at the adset level, we have pants adsets, tops adsets, dress adsets, and that's where the segmentation of categories occurs, and we make sure it's an so we can control budgets here. Or option number two is we keep this structure, but we just consolidate all the regions into the campaigns. Now, why or when would you want to do that? It's if we are shipping the UK orders from an Australian fulfillment center because in that case it doesn't matter

other than our targets and our goals as a business whether the UK revenue goes up, goes down, goes sideways because we don't have any holding costs of inventory in the UK market. All of our stock is held in Australia. So we just need to sell through the stock regardless of where we're actually selling to. Now, there is a bit of complexity here because it's fashion, which is that Australia is southern hem, UK is northern hem. And so, they're going to be different seasons. And so, what's selling well in Australia is not going to sell well

in the UK and vice versa, unless there's transseasonal products. And so, in this case, you are probably making purchase orders for the UK season, in which case you do need to move inventory. So, there's this nuance in understanding what does this region actually look like within the business? Are we holding inventory? Is there risk? do we need to hit particular sales targets because that information is then going to infer into the account structure which is why understanding the business and the complexity and the goals is so important when we're trying to translate into a structure.

I'm likely going to go with consolidating this down having the segmentation at an adset level and we have an Australia cold campaign and then We have the same thing in the UK. The reason being as well is that the UK likely sits on a different subdomain and so we need all of the UK ads driving to different URLs than where the Australian ads are driving. And so because of that, we need the separate campaigns and therefore we're going to move the category segmentation down to the adset level. Then in fashion, what does do incredibly well

is dynamic product ads, particularly in retargeting. So, we're going to throw in a DPA retargeting ad that's going to go to engaged audiences, which is just like your 90-day website visitors. And then we're also going to throw in a final campaign, which is your retargeting on existing customers, which is going to contain two adsets. We're going to have new arrivals in there. So, we're constantly putting new arrivals in front of the existing customer base. And we're also going to have a DPA. We're going to have it as two separate adsets because if we consolidate all

the spend will just go to the DPA and we'll never be able to actually serve new arrivals to existing customers. These campaigns are obviously because they're cold going to have existing customers excluded which is why we need this. Um targeting existing customers in fashion is also much more incremental than any other category that we've seen. So it actually is worth the spend allocation which is why you have on the surface two brands at the same revenue level. But when it then comes to understanding the business and all the complexities, the gross margin profile, the personas,

the countries, the skew count, it then translates into a very different account structure. All right. So, if you're spending sub 50k a month, I'll give you A generic structure that most people can get away with. Will it be ideal? Will it be custom to the business? No. I've gone through all of the reasons as to why you should ask yourself those five diagnostic questions to be able to better understand how you should structure yourself. But if you just want something generic, if you need to throw something up, here's what you should do. Number one is

a dedicated testing campaign. Now, at this spend level, this is probably all that you need. Now, what this actually looks like is either an or a CBO, and we'll go through later which one you should actually choose with adsets sitting under it. Three to five creatives per adset. Now, you can go way more than this if you have the creative volume to support it. Just most people at this spend level probably don't. And the creatives are being segmented based on concept. Now concept for people who haven't watched any other video of ours is the intersection

right here in the middle of an angle, an offer, and a persona. And so you're putting a persona, angle, offer together, and then you're making creatives under that. And that sits in an adset. When you then go and make more creatives for the concept, you have an option. You can either launch it in the existing adset or you can launch it in a new one. In terms of which you do there, if the adset is not performing, throw it in. Try to get the adset to perform. Throw more creatives at. If it is performing, if

it's hitting KPI, if it's doing well, don't touch it. Number one rule of meta media buying is never touch something that's working. If it's working, don't touch it. Do something next to it. Don't ruin the thing that's working. So, if it's working, don't touch it. Launch the concept as another adset. So, you might end up in a position where you have 10 adsets and six of them are for just one concept and it's a bunch of new creative that you've made over time and everything's working because that ends up being the case is One concept

ends up outshining everything else. Now, it's not that important at this scale of business, but you do want existing customers excluded. We don't want our testing getting skewed around based on existing customers coming through a few ads and making something look better than it actually is. You also want to make sure that your attribution setting is on 7-day click so that you have data integrity in the numbers that you're reading. Or else, you also end up in a position where some things might look better than others just because it's claiming conversions that has nothing to

do with it. Then, you can layer on a scaling campaign. Now the idea here, the premise of having a scaling campaign is that one of these adsets performs well. So let's say this adset at the top is performing well above your target return of 5x and all the other soft metrics of really good too. You start scaling this up. You put more spend more spend over time and you're ramping it up and then eventually return on ad spend drops to a point that is unprofitable and so you need to pull back spend. Now, when you

do that, you effectively find an equilibrium where you find the daily spend level where that creative can keep sitting there, continue churning over results that you're happy with. You can't push it any further. Anytime you do, results drop. So, you have to come back. And because of that, you're kind of not happy with it. You're like, "Ah, we could only take this creative or this set of creatives to $300 a day." But obviously, I want to spend way more and I want to scale. So, what do you do? It's at this point where people often

turn and go, "Well, we can't scale this adset anymore. It's not working. So, how can we take these creatives and just launch them elsewhere within the account structure to be able to get more spend through them? And that's effectively where scaling campaigns are born, which is you take creatives that have been maxed out. You can't get any more spend through them, and you go and just dump them in a new campaign. And you hope that the new Campaign can get even more spend through that ad. And a lot of the case, it can't. For whatever

reason, you take the creative, you launch it in a new campaign, you keep this on, keep this spending, but you go and launch it elsewhere, and you can likely get a little bit more spend through that ad. And so rather than the ad within the account holding $300 a day, maybe you go at this point and you layer it in on a scaling campaign and a scaling campaign can get it to hit. And so total spend across the entire account on this asset is now $400 a day because it's being propped up by the scaling

campaign. That's the idea of why you've launched a scaling campaign. There's absolutely no reason to have one if you don't have ads that are working well. like there's no point in adding this complexity unless you have creatives that are doing well that have been cranked to their maximum potential and then you're just trying to squeeze even more out of it. A big mistake I see is that people will have something that works and they get it to like $60 a day. So like a meaningless amount of ad spend and then they'll go, "Oh, we can't

push it any further. Let's roll it into scaling." This is nowhere near enough spend to be able to start trying to scale the ad even further. Good way to conceptualize this is that this media buying move of taking a winning ad and then trying to get more spend into it using an adjacent campaign will get you on a good day an extra 20% daily spend through the app. So if you have an ad that's spending $60 a day, this kind of move in the account is going to get you like an extra $10 in daily

spend. Not worth it. Just make better ads. Fix other things in the business. There's probably a landing page issue. There's probably an offer issue. You still probably haven't found a concept that's working. There's like a million other things to put your effort towards rather than trying to media buy Your way to an extra 10% of spend on a really low base. If your ad spending $400 a day actually worth just taking it and putting it in a scaling, it's a relatively loweffort move. It doesn't take much time and yeah, you're going to unlock an extra

$80 per day in daily spend. If you do this across like 10 ads all at once, you take 10 of your highest performing ads and you dump them in a scaling campaign, it might allow you to unlock an additional $1,000 day spend. And that actually is worth it. That's a decent unlock. it's worth having this increased degree of segmentation and complexity in the account. And then lastly, you have the retargeting campaign. This is absolutely an optional campaign. When we talk about retargeting, there's two different types of retargeting. There's website visitor retargeting. So, effectively warm audiences,

people that have shown intent, they've taken some kind of action, but they haven't purchased. And then we have existing customer retargeting. So, this is someone that's actually purchased from us. They're all the way down at most aware in the stages of awareness cuz they're a customer and we're trying to effectively get them to buy a second time or a third time. Now, for most accounts, 90% of them, you don't need a website visitor retargeting campaign. And the reason being is that the testing or the scaling or whatever currently exists will allocate a good amount of

spend anyway towards retargeting website visitors. And you see this by doing an audience segment breakdown within the account. So, if you hit breakdown, you hit audience segments, you'll be able to see as long as this is set up within your advertiser settings in your audience segments. So, make sure that's set up. You'll be able to see spend, rorowaz, etc. on new audiences as well as engaged and then Existing. Now, you shouldn't have any spend to existing cuz they should be excluded, but you'll be able to see your spend towards engaged, which is your website visitors.

And there'll likely be enough spend here at a high enough frequency that you don't need a dedicated campaign to force more spend through that elites. It's okay on existing purchases because you have it excluded here. You probably want some spend towards existing customers. It just depends on the retention dynamics in the business, the category, and the stage of business that you're at. If you're a new business, which is sub 50K a month, you probably are. You're also probably a small business. You don't have that many existing customers. And then depending on the product portfolio and

the retention portfolio, so is this furniture where you're like going to have no repeat purchases really. It's very low. It's very infrequent. or is this supplements low average order value which has super high repeat rates in its subscription model. Okay, well different story. We probably want some spend there. So you need to contextualize this to the actual business model and the expectations on repeat purchasing. If there's an expectation of high repeat purchasing, put some existing customer spend in there. Keep it low, keep it controlled. Look at frequency as the measure. If you don't expect repeat

purchases, don't have. Now, the reason why this structure works at this spend level is number one, it's consolidated. You probably have not a lot of conversion volume at this spend level. And so a product of that means that you shouldn't have a lot of segmentation. You want all the conversions consolidated up. You need structured creative testing at this spend level because you probably don't have anything that's a very large winner or else you'd be spending more. And so you want an structure where you can force spend through adsets and start to learn stuff. And then

number three, at this spend level as well, you probably don't have a large product portfolio. You probably don't have too much complexity in the business model as it stands. And so we can keep things relatively simple. In terms of KPIing this structure on the testing campaign, you want to KPI at the adset level. Did the adset hit our target cost per acquisition or our target return on ad spend over a roll-in window, which is reasonable. Okay? And what window in which you read data is dependent on conversion volume. So you can't look at three-day windows

if you're only doing three conversions in a 3-day window. You can look at two-day windows if you're doing a thousand conversions a day. So, it's all based on contextualizing uh the windows that you're looking at for performance based on how much volume the business is actually doing. At this kind of scale, I wouldn't really be reading data on any shorter than a 5day period. And then if the adset is hitting KPI, budgets go up in 20% increments per day. If it's not hitting KPI, we diagnose why. What do we think was wrong in the creative?

Let's then go and make more ads, more concepts, and let's either wind that adset down or turn it off completely. On the scaling campaign, you want to just be KPIing all the way up at the campaign level. There isn't any adset segmentation on a scaling campaign, at least at this level. And then on retargeting, you want to be KPIing based on incremental return on ad spend by doing an attribution breakdown. And also frequency shouldn't jump to above a seven on a 30-day window. The four common mistakes at this level of spend is that people test

inside the scaling campaign. There's no exclusions on the scaling campaign. They kill testing way too fast. So the window in which they're looking at conversion data is way too little. And then they set budgets on the adset level based on the 50 conversions in a 7-day window rule, which ends up With just way too much budget for their particular business. So you actually need to back propagate budgets at the adset level based on your average order value and expected CPA. So let me quickly break that down for you so you understand what kind of budgets

you do need to be setting on the adset level. So when you're looking through adset level spending, there's really two factors that you need to think about. You need to think about time to outcome of the test and then you need to think about total budget required for the test. So what does that actually mean? Well, let's say that you have a target cost per acquisition. So you want to be getting customers at $50 per customer. You then come up with a new concept. You make 10 ads under it. We then go and launch it

at the adset level. What we really want to figure out as quickly as possible is is that concept working or not? Now we need to set a barrier or a threshold for how much spend are we going to put through those ads through that concept before we decide yes or no. Now the general rule of thumb that almost everyone uses is you take your target cost per acquisition. So this is our target. You times it by three and that is how much you should spend before you decide whether to keep the ad on or off.

If you get three conversions in this time, if you get four, even better. Keep running it. If you get two, uh, yellow light, let it run a little bit longer. Let it run another $50 and then we'll decide. If you get zero or one, cut it. Now, that as a rule of thumb is pretty decent. Now, I would always let it spend a little bit more. I'm always willing to be a little bit more gracious because I know that the initial spend is really at trying to learn and figure out who to target. And then

once it figures it out, the ball gets rolling and you actually see your efficiency client. And So, I actually prefer this rule if possible. And if the client's okay with it, they're more of a times five. So, we wait till we spend 250 and then we make a concrete decision. There's no yellow lighting. It's either a yes or a no because we have enough spend volume here. Okay, cool. So, let's say that $250 is what we want to spend. So, we know the total budget of the test. We're going to spend $250 on these new

creatives before we decide whether they're a success or not and whether we turn them off. Then, we have time. This is how quickly do we want the test outcome? Because technically, we could just set $250 a day as the budget and we'll know whether this adset in this group of creatives has worked today. We'll know by the end of the day. Now, the issue with that is that there are a couple components that will play into the time duration that you need to allow a test to run for. Number one is daily seasonality. The reality

is is that there is going to be some days of the week in which you do better and some days of the week in which you do worse as a business. If you go and launch a test on a good day, you will see better results. If you go and launch it on a bad day, you will see worse results irrespective of the actual test variable, which is the creative. And so if we run tests too quickly, we don't get the ability to encapture the whole seasonality of the week and therefore we can get a

biased view of performance. The second thing that impacts time is time to purchase. And so for a lot of brands, people don't instantaneously purchase the first time they ever see an ad from. Okay? You don't just serve an ad to someone and they go, "Oh, great buy." Normally it takes two impressions, three impressions, four impressions, Maybe a couple clicks, and then over the course of a 3 to 4 day period of warming that user up, then they make the purchasing decision. Now, if we run this test in one day, well, guess what? We're going to

show the ads to a bunch of people. They might go, "That's amazing. These ads are great. I'm now interested." But then you kill the ads, you turn the test off, other ads go and target those users in retargeting. They all end up converting to other creative, but you killed the primary top ofunnel ad that you want to test it because you only gave it one day. And so you need to not only give it a little bit of time because of daily seasonality that might skew the results, but also because people take a little bit

of time to buy. True. And this applies not only to Facebook, but Google and all the other platforms as well. So because of that, we need to balance these two variables. We need to figure out what is the budget for the test. And then how much time do we want as a feedback loop? Ideally, as fast as possible. We want to know the outcome quickly so we can continue to iterate because ultimately the speed of growth in a business is a function of the speed of cycles of feedback and learning and iteration. And so we

want to iterate and learn as fast as possible so that we can grow. But we want to give it enough time that it encapsulates daily seasonality and time to purchase. What does that often end up coming out to? One week. One week is generally a good duration to run a test for. Now if you're a big business with more stability with not a long time to purchase and cut that down to three to four days. If you're a tiny business with tiny budgets with a incredibly long time to purchase and massive daily seasonality, well, okay,

maybe we need to extend that to 10 to 14 days. Okay? So, all once again, always going to be respective to your particular business, which is why in this whole video and all the content that we put out, we talk in frameworks rather than actuals. Rather than saying this is exactly what you should do, we say here's how you should think through the problem. Here's the two variables that impact it. Now think through the problem yourself in your own context and then come up with your own solution because every solution, every budget, every adset is

different for every single business. Now as you jump to 50 to 250k a month in ad spend, if I was to prescribed you with a campaign structure, which once again we don't like doing cuz different based on all the commercial objectives of the business, but if I was to, you still have the retargeting campaign, this is likely still going to be just existing customers. Very rare you need to layer in website visitors here. You might, probably unlikely. And then in terms of the cold campaigns, this just builds out a little bit further. So you're going

to end up with probably more adets at the adset level cuz you're going to be doing more creative testing and likely have more creative volume, which is going to introduce more adset volume. And then potentially you're also going to extend out to two maybe three campaigns. Now, if we're looking at this as if it's a CPG brand with one skew, this campaign segmentation is actually going to come from different funnels and different approaches. And so you might have a funnel or a persona or an approach depending on the vernacular that you want to use to

explain it. But you might have a funnel that is targeted at old 65 plus year olds for arthritis, right? And this is a supplement that we can pivot for that. In fact, let's just call it fish oil, right? Fish oil applies for old arthritis. All the ads as well are going to be mirroring this type of style that works with this age demographic, which is going to be VSSLs. It's going to be native statics. It's going to be testimonials from an older demographic. And then actually probably throw on like TV style ads as well that

you're chopping into UGC format. Then you might have the product repositioned. All of the ads are now for young people 25 to 35 focused on like brain health or performance at work or something orientated around that. Now you could also get this funnel working. The landing pages are going to look very different because you want different type of demographics and photos on the landing pages. Um the ads are going to look very different. Probably the profiles that you even run this through might look very different as well. This might be heavier on whitelisting. This might

be on the native page. But because this is almost two different businesses in itself because the funnels look very different as a product, this is two separate campaign. So that's what that would look like in CPG with one skew. If we're talking about like fashion with a bunch of SKs, this just ends up being some kind of category segmentation. It might be new arrivals segmentation. So you have your normal existing cold campaign that you had at lower spend, but now you just go and bolt on a new arrivals campaign um for the sake of pushing

new arrivals and having higher sellrough rates here. This might be a particular product category. When we do an LTV analysis, we find out that this particular product category, let's call it pants, ends up with not only a higher average order value on first purchase, but much better repeat rates and retention. And so as a product of that, we make a second campaign just for pants where we want to have a higher cost per acquisition KPI. So we can be more aggressive in acquiring customers through here because we know that our pants category acquires really high

quality customers. And so as a product of that, we get a second campaign. So generally when you're going from the sub50k range and then bridging into 50 to 250k. What ends up happening if you're a good media buyer or a good Performance marketer is additional campaigns will start getting layered in that's has a very specific reason or commercial objective that aligns with the business's fundamentals. If you're just layering in more campaigns for like the sake you're like ah we're spending 100k a month right now. One campaign seems weird. Let's do two let's do three. you're

breaking the fundamentals that I went through before, which is that every segmentation decision needs to be done with the ability to read data and make decision loops through an underlying structure that has data integrity. So, if you're just introducing segmentation for the sake of it, you're breaking all the rules that we've gone through in this case, you want to look at segmentation, but only if it makes sense to better align with the business's objectives. The main key at this campaign level and where the real skill unlock is and where some people just smash straight through

this spend level and some people stay here for a very long time is in continuous testing of different concepts so that you can find the 20% of concepts that are going to do really well and are going to be able to hold hundreds of thousands of dollars a month in ad spend. Now we have a YouTube video called meta ads creative strategy in 2026 the full system where we spend literally 30 minutes talking about how to create concepts. So, I strongly recommend you go and watch that video. But if you're unfamiliar with the concept, one

minute run through. You want to be building out personas, angles, offers, and then ad types, building this matrix out, and then having your ad sets reflect this structure, and then continue to test ads under each concept. An example of this, if you're selling teeth whitening, is that a persona could be coffee drinkers. The angle is whitening without sensitivity issues. The offer is just the product. There's nothing special there. And then the ad type is userenerated content. And obviously under this ad set, if it's coffee drinkers, whitening without sensitivity, we can rotate in tons of different

ad types. Like the ad type is flexible. This is infinite. The next angle is brides. And you go, this is very different, right? Coffee drinkers into brides. How do brides relate to teeth whitening? Well, we can frame this around get ready for your wedding in 14 days. We can do a bundle offer that's specifically curated for fast 14-day turnaround to get white teeth for the wedding. And then we can do this to a testimonial of an actual person that just had a wedding in the dress. They can do a before and after. So they can

document the whole process. And the real key here and where this really accelerates is that once you get something that's working, let's say this bride idea does well and this ad does really well. Well, you double down. You start making a ton of creative around this concept and you have a custom landing page that has continuity through it. So we might actually spin this bundle off and call it like the wedding in 14day bundle. have an own custom landing page, have all these testimonials on the landing page, build the entire funnel around this angle, and

then this can scale to 2, 3x the volume, and we can really saturate this market while at the same time continuing to work on other angles and other persona. And the last one I have here is an obvious one, which is smokers to remove stains. You give them a subscription offer because they're continuing to smoke, so they'll have this issue forever. And so you want to put them on an offer that's relevant to that. And then this is through founder head talking content because maybe the founder is a smoker and that's why they started the

business. So, the reason why concept level testing becomes so important at this tier of spend is because each campaign should be generating enough spend volume, enough conversion volume to be able to support a degree of segmentation at the adset level. It allows you to then start to KPI based on concept rather than just the campaign as a whole, which allows for better directional feedback in creative. And this is really a core key here, which is that this creates a creative strategy because you have the Strategy, you come up with all the creative, but then most

people just throw it in the account, and then they just make more creative, and they throw it in the account, and they make more creative, but they don't actually look at the account and go, "How is stuff performing? How are we reading this data? How are we drawing insights?" And then, how is that informing the next batch of creative that we're making? And so the creative refreshes become targeted, the losing concepts get more attention, and the winning concepts get more volume. You can also then begin directing resources upstream to landing pages, retention flows, product portfolio

expansion based on concept level testing. As an example, let's say that the bride's persona starts performing incredibly well. What do we do? We increase creative volume. We create a custom landing page. We change the offer accordingly. and potentially as a byproduct of changing the offer. We expand the product categories to meet this audience as we're generating so many existing customers through this funnel. And so we might have a retention win back flow off the back of this that is post wedding. We send them a free gift saying congrats on getting married. Here is an additional

offer for you that pulls them into a different product line that meets them at the stage of life of where they are. So you can start to get really creative in terms of the backend retention, flow, sequencing, and product portfolio based on direct feedback of what type of customers are we acquiring and from where. If you end up keeping your concept targeting super broad and you don't niche down in this way, you don't get accurate persona insights and that then bleeds into inefficiencies in the rest of the business. So this is really also telling you

who your customer is. Who is the customer? Who are we acquiring? And in what percentage Allocations are they? Are we getting 40% of our customers because they're smokers? 40% because they're about to get married, etc., etc., and then we can start to craft the entire business strategy around this. The last reason why concept testing becomes critical at this scale is it supports a large volume of ads. So, at this spend level, you should be launching really between 100 to 300 new ads per month if you want to grow. If you don't want to grow, don't

launch that many ads. But if you do want to grow, you should be launching around about this volume. This volume becomes actually quite sustainable for entering into an account structure if you have built out to let's say two campaigns with 10 adsets within each campaign because that's 20 adets which means on the low end you've got five ads per adset. On the high end you've got 15 ads per ads set. Very reasonable. In fact, you could actually have less campaigns, less adsets and you're fine here. The ad to adet ratio is completely manageable. So what

then happens at 250k plus? pretty much everything at 50 to 250, just more of it. And so when we come to campaign segmentation at this budget, and this is a enormous budget range here, cuz I'm saying 250 all the way up to $20 million a month in spend. Okay? So it's an enormous range and therefore there's an enormous degree of complexity difference. And so because of that, unlike the other two sections, I'm not going to give you an exact account structure that you should run because it would just be ridiculous. It won't be applicable to

anyone at this spend level. and instead I'm going to give you a bunch of strategies, a bunch of advice, a bunch of tips, and then I'm going to run through an example so that you can get as much value as possible on how you should be structuring at this spend level. Number one is you do want To potentially consider layering in a second ad account. Now, this strategy changes a lot and what I would have said 6 months ago is different from what I'll say today and what I say today will be different in 6

months. And so, I don't want to give you too much tactical application here because it'll just be outdated. But the idea here is that you run the same pixel but you run different bid logic. So if you're running maximize conversions in the main account, you might run maximize conversion value or you might run cost caps or big caps in this account on the same page. It could be different page as well. And the idea is that you'll start winning different auctions from this ad account that the primary account just isn't bidding on. And so it

will allow you to get more volume through effectively the same creative and the same pixel and the same page, but you're entering different auctions due to it being in a different ad account. The disadvantage that you could argue is it might increase CPMs and ad costs for you because you might be cross bidding against yourself. Now, if you're using the same page and the same pixel, you shouldn't really cross bid that much, but it's definitely an argument and no one has really been able to prove whether this is the case or not. The main other

obvious advantage here is that it derisks the business enormously because if you have a second ad account and the primary ad account gets banned or the billing method goes down for whatever reason, you don't just lose all your new customer acquisition in the business if you're overrelyant on meta ads. Instead, you have two ad accounts. One might only hold 10% of the spend of the other, but if the main one goes down, you can just crank up the second one, and it prevents a doomsday scenario where you might not have any revenue for a week.

Number two is that at this spend level, you absolutely need a page strategy. As I said earlier, you're going to hit your ad limit without a doubt. This is a guarantee. And so, what is the strategy to be able to avoid that from happening? And I gave you the three options earlier in the video. Number three is that you will generally start to get complexity getting introduced in more media buying tactics at this level of spend. And this is the level of spend where it actually does start to make sense to be playing around with

the 3 4enters because 3 4% when you're spending a million dollar a month is actually quite mature and it could add an enormous amount to bottom line profit. And so having someone dedicated on trying to media buy your way to more efficiency is genuinely worth the investment. This is where like bidding complexity will start to be introduced. You're not just running maximize for conversions in the account across everything, but you might start introducing bid caps or you might start introducing cost caps or you might have maximize conversion value or target rorowaz that's sitting next to

these campaigns. And the reason being is that each different bidding strategy will enter auctions differently with different bids and you will generally win more auctions as you start to diversify the bidding strategies within the account. The reason why I almost never talk about this in any content is because 99.9% of people are not spending over 250k a month and so should not be concerned whatsoever with bidding strategies. Now counter to that 60% of our client portfolio spends more than 250k a month. So for us internally the bidding mix actually is a big deal and this

actually is something that we need to think about and think through but for most people ignore it. What will also be the case at this spend level is you'll generally have some kind of international market expansion in which case you need to start thinking through the complexity that gets introduced into the ad accounts from that international expansion. Whether you run secondary ad Accounts for different regions, how you deal with pages, how you actually deal with the backend on Shopify or however you're hosting the different regions. A general word of advice is that I would advise

against segmenting countries out at an ad account level. And the reason being from an agency perspective is that it will increase your costs because working across multiple different ad accounts increases labor dramatically. And so I would always rather work on one ad account than working across seven ad accounts which we have some clients that have seven ad accounts for seven different regions. And it adds so much additional labor and complexity into the management across it. Now the reason why you would have all those ad accounts is really only one reason and it's that you want

to get build in the local currency of that region. So if you have a US web presence, you want the ad account to bill you in USD because maybe you have a US bank. If you have a UK presence, you want all of your money flowing through in great British pounds and so you need a separate ad account. That's really the only convincing argument I have seen for introducing segmentation. Other than that, everything else is solvable. You could say, "Oh, reporting is better because we can plug these ad accounts into our dashboards and reporting." Yeah,

but you can just add filter rules based on country segmentation, or you can just add filter rules at a campaign name level, and all of that's solved. Like, you don't need to introduce ad account complexity other than for the reason of just getting built-in local. Now, let me give you a worked example of an actual real ad account that's spending 25K per day, which is about 750K a month. So, closing up on a million a month, and this performs incredibly well for them. Now, note this will not guarantee results for you because your business has

its own complexities and its own differences and You should think through everything that we've gone through so far in terms of how to structure it. Or you can obviously always click the link in the description, reach out to us. We will do a free audit as long as you're doing at least $5 million a year in revenue and we can walk you through what that account structure might actually look like and we can provide it to you. So, we have a testing campaign at the top. It's an So, all tests are still being done at

the adset level. I have actually seen accounts that are spending $300,000 a day in budget and they're still running tests. So I see a lot as push back, oh you shouldn't run testing once you're actually a big account and you're spending a lot. That's not true at all. Okay? Like you can run CBO or It's up to you and there's benefits of each one. It just depends on how you want to manage the account and also what the particular nuances are of that business. ABOS at this spend level work and you can perform incredibly well

and there's reasons why you want to do them because you want to force spend through new tests. and CBOS can also work incredibly well at this spend level. So this is once again up to you, but this ad account runs a testing campaign as an Number two is then a scaling campaign, which is a CBO, one adset, cost caps. Every 1 to two weeks, the top performers in the the post IDs are taken and they're launched into the scaling campaign. Now, they're not turned off in the testing campaign. The testing campaign is also used to

scale at the adset level. There's adsets in here spending multiple thousands of dollars a day. You still scale in the testing campaign. It's just this is a strategy to try to squeeze more out of an existing post that's doing well. Third campaign is a promo campaign. This particular account runs promos every three to four weeks. And so as a function of that, we want it segmented out. Why? Because the promos turn over a lot. And so if you're launching promos in the testing and the scaling campaign, it will disrupt the learnings of the Campaign because

you're constantly just turning stuff off and launching new stuff in it. You want to leave stuff that's working and you want to add pipes at this level of spend. So we want to add this in separate to not impact the performance over here. This will also generally go after a different customer. So it is a little bit different in prioritizing price sensitive consumers and so as a product of that we do want it to optimize a little bit separately. The fourth is an advertorial campaign. So advertorials as an additional funnel for this business started to

do very well in the testing. It started to do so well to the point in which it was consuming about 30% of total spend. And so it actually made sense to just pull it out and have it dedicated here. So it can be KPIed on its own and it can be looked at as a completely separate funnel within the account. And then lastly, a DPA campaign. This is primarily for retargeting existing customers as well as a little bit of website visitors. This is running on incremental attribution so that we're attributing correctly based on its actual

incrementality. And this is kept at a relatively low spend in line with frequency. What doesn't change at this spend level? What stays the same? Well, number one is the two non-negotiables. Everything that we introduce, every extra bit of complexity needs to still ensure that we have data integrity so that we can read the data and then make decisions. And it all needs to have commercial alignment. Is this aligning with the products, the margin, the portfolio, the personas, or are we just adding complexity for the sake of it when we don't need it? Number two is

the breakdown effect still applies. Suddenly just doing breakdowns or looking at the ad level does not become more productive here than it does at a lower spend. It is the same thing. The principle still stays. Number three, the concept Framework of launching creatives still remains the same. You can have this framework up at 300k a day and spend. In fact, I would recommend it. Normally, the big accounts that are spending those levels are structuring creative and testing it in this way. And number four, you want to be KPIing at the ad set and campaign level.

This doesn't change. So the premise really is that you're adding structural layers into the account as spend increases and business complexity increases, but you're not giving away the fundamentals that we spent the first 30 minutes of the video setting in place. And then the question I've alluded to throughout the whole video is versus CBO. It's the wrong question. It's honestly personal preference. And the reason being is that this is just what your risk portfolio is. Okay? If you want more risk in the account, but you want better efficiency, you go for CBO. The reason being

is it's going to distribute more spend to the highest performing ads and the highest performing adsets, which is good, right? You're going to get the highest ROI or return on ad spend in the account. The disadvantage is that all the spend is just going to go to the highest ads and so you can launch new tech, new ads, and all this money on creative production, put it into the account, and it gets no spend. So, if you just went and spent $20,000 on a new campaign shoot, and then you went put all that content into

the account, it's not spending. What do you do? or let's say that you went and just spent $20,000 on a very expensive influencer to run a partnership ad for a 60-day period, put it in the account, isn't getting spent. Now, you can put minimum spend caps in CBOS, of course. So, you can go and put minimum spend limits. You can also put maximum spend limits as well, but then you're effectively just running an where Either you're just increasing the complexity of management because running min and max spend caps is just annoying. It's just a harder

management tool within the account. Or you do this with like let's say 60% of budget and then you let 40% of budget get distributed by meta how it want. Now that's actually in my opinion the best structure to run. Most of the structures that we run these days is that structure which is we're running CBO but we're using minimum spend limits and maximum spend limits but we're allowing about 40% of total budget to just flow wherever it wants and 60% we're tightly controlling and saying you have to go here, you have to go. The reason

why I'm not a proponent of this at like lower spend level accounts is that this is just like a lot of management and it requires a lot of oversight, which is fine for us when we're managing very large accounts with large ad spend with large revenue. It's worth your investment for someone like us to be able to do this kind of degree of management and budget segmentation. But if you're managing like 10k a month in ad spend, this is it's just kind of overkill. Like just run an or just run a CBO. Choose your risk

versus efficiency profile and run it. So that's really the thought through of the process. So if you prefer ABOS and once again I have seen ABOS running with majority of ad spend on ad accounts spending a4 million a day and I've also seen CBOS running on ad spends quart million a day. So it really comes more so down to personal preference as to how you want the testing methodology rolled out in their account. Do you want spend forced through every new creative or do you not? Do you trust that the algorithm is only going to

give spend to an ad if it's going to do well? That all comes down to your trust and the algorithm what you have seen historically subjectively within the account. Has the algorithm not given spend to an ad Before? And then you went, "Oh, I thought this is a good ad. Let's put a minimum spend cap." You put a minimum spend cap, it becomes a winner. Immediately, you start losing your trust that the algorithm knows what it's doing cuz it wouldn't have spent on that ad if you didn't force it to spend and now it's the

top performing ad in the account. So, it's going to come down to honestly your own bias and subjectivity in relation to how much you trust the platform in whether you're going to go or CBO. And I honestly don't think there's a right or a wrong answer. You can run whatever. They both perform well. The only caveat I will add is what I said previously, which is that if you have really high AOV and low conversion volume, I wouldn't recommend CBO. And the reason being is that the CBO will prioritize soft higher intent metrics like clickthrough

rate and CPCs. And so you'll just end up with spend getting distributed to the best soft metric ads rather than to the best ads that are actually going to drive conversions within the business. There is also the complexity of we will move out and in of ABOS's and CBOS depending on seasonality which sounds kind of weird but because if you put these names aside and you just think high risk high ROI and then you think low risk slightly lower ROI like the ROI isn't that much lower here but it's a little bit lower let's say

10%. Well I want to pick this during Black Friday and November December and peak periods where we're just trying to spend as much as possible in a short window. I don't care what creatives get spend. I just want to maximize efficiency and maximize spend. But then in like Jan Feb in Q1, I kind of want this, right? I want to set Q1 up for a bunch of creative testing. I want to test as much as possible. This is a great environment for testing because you aren't artificially inflating conversion rates and making ads seem like they're

doing well when they're actually not. We have a relatively low Risk profile during January. We don't want to increase our risk profile during this part of the year. And so what we will often do is actually rotate into a CBO in Q4 for some accounts and then roll it back to an in Q1. Now regardless of whether you run ABOS or CBOS, the ad set composition remains the same, which is that you always want a minimum of three ads in an ad set because it allows the adset to sequence across ads. If you only put

one ad in an ad set, it can't sequence across anything. And so it's just going to serve that one ad to users, which ultimately isn't going to get people to buy because generally people need to see an ad 2 3 4 5 six times before they actually purchase. You want a minimum of three ads. You want a maximum of really infinity depending on spend. And so you can go all the way up to the maxed ad limit at an adset level. That's fine. We have accounts where we run that. We might have 200 creative in

an adset. Generally for most people watching this, you don't have anywhere near enough creative volume to support 200 ads in an adset level. Like you may as well have a degree of segmentation there. and there's likely going to be different concepts in those 200 ads you don't want to consolidate anywhere. So, as a rule of thumb, you can think about three to 15 ads per adset. Number two is that you want all of your ads resonating with the same audience. Now, this doesn't mean that you need different stages of awareness or different types of ads.

So, this is a really common question I get, which is that if I launch an adset with five ads in it and it's all under one persona talking to moms 40 to 50 who need a rain jacket, do I put every stage of awareness in there? Do I put my super topfunnel VSSL style ads that are trying to sell this weird raincoat through a story line and a 5-minute video, but then also just put a static ad that just says here's a RCO 50% off. Do I put them in the same ads there? Because they're

at very different stages of awareness. So, do I split that at the adset level? And the answer is you can do either. It's just dependent on how you want to read the data in the platform. And so, let me walk you through an exact example. So, option one is you split it out. You have one adset which is the persona, the raincoat, but this is very top offunnel creative. It's high up on the stages of awareness. And then you have adset two which is going to be very bottom of funnel in the stages of awareness

in the way that this creative is designed. That's adset two. Option two is we just consolidate it. This has our top of funnel, it has our middle of funnel, it has our bottom of funnel ads all consolidated under the one adset. Which option is better? My understanding of meta's machine learning and the way that conversion data is split across the account generally pushes me a little bit more towards option. The reason being is that we want these ads to be able to easily sequence against each other, bring people through the journey and convert. However, I

have seen option one also work. The reason why option one is an inferior option. The reason why I would recommend most people don't do this and they go for option two isn't to do with the actual machine learning and the way that it works. Instead, it is to do with your own human bias. This adset down the bottom here will have a 4x row. This adset up here will have a two. This will have a six. Now, if you go into this account on option two and you look at this adset and it's at a

4x, your conclusion is this concept is doing well. Let's do more like this. Let's also scale budget. Incredible. Now, let's say instead you go into this ad account and you go, okay, we have two adset. This one's not doing well. This one's doing well. What do we do? Do we blend the numbers and say that overall the concept at a 4x therefore do more Top offunnel creative? Is the top ofunnel creative really performing well? Do we turn this off and just like leave bottom ofunnel creative for this? Like what do we do off the back

of it? Now the answer in terms of what you do is you should just blend the numbers. Assume that these two adsets are working together to get the conversion and therefore do more scale it etc. Exactly what you would do here. But due to your own human bias in the way that you're going to read the data, you're inherently not going to want to do that. You're going to look at this 2x and particularly if you're working with clients that aren't inclined to think of it like this, they're going to go, "No, no way. Don't

put more spend into a 2x. That's not a good idea. Put more spend here. Even though this is obviously bottom of funnel and then this traffic is driving into here which is converting and this is getting the last click attribution. And so this just adds a lot more complexity into the human decision-making process. And the human decision-making process is where this falls apart. And so I would recommend that you just put your middle, bottom, top of funnel ads all together under the one adset. Now this is an underrated topic when it comes to adset structure,

which is actually naming adsets correctly. What you'll see in most ad accounts is something like this. March 7 creatives. So, as batches of creatives are made, they're just named, thrown in the adset level, and right. The issue with this is that there's no ability to easily filter into what concept we're actually testing here. This also infers that we're just batching a bunch of concepts together rather than carefully structuring them so that we can get clear data insight and iterate into the creative team. And so, as a product of that, when we pick up accounts like

this, it's very annoying and complex. And often we'll actually go back and do retrospective renaming of old ads set so that we can at least get some kind of clear data structure historically so we can infer what best to double down on moving forward because if you don't have clear naming conventions in place of for example you'd want to rename this into the exact persona angle offer and then content type if you're going to group a content type and then date of launch so that we can start to bundle and KPI different concepts across a

campaign. So what about when it comes to retargeting DPAs and existing customers? Here's three principles you need to keep top of mind. Retargeting is a capped supporting layer. It's absolutely not a growth driver. This is unfortunately a mistake that I see in so many ad account audits that we do, which is that you go in, you do an audience segment breakdown, you pull up from the bottom of the account, there's a little button there, and you can just see total spend distribution across the account. And you just see like 50% of spend going to existing

customers and engaged audiences. Now, like 50% of your spend is going towards support spend. that's not actually going to drive incremental new customer growth within the business. It's just supporting existing spend and it's not even that incremental. And so the key premise to keep top of mind here is that you want to be at all times minimizing retargeting spend as much as possible. This is the different thought process between cold and retargeting which is on cold on new customer acquisition. The constant question that we're asking ourselves internally is where can we spend more money? What

platform? What channel? What campaign? What ad? Where can we spend more? We need to spend more to get more incremental lift as long as it's profitable. As long as we can make a profitable exchange, we can put $1 into this ad and get four out. Or we can put $1 into Tik Tok and we'll get four Out. We want to keep putting more money there to grow new customers and grow the business. On existing customers and retargeting, it is the opposite frame. We want to be thinking, how can we spend the least amount of money?

Where can I take money out of? Are we spending too much retargeting on TikTok? Take it out. Take it out of Pinterest. Take it out of Meta. Because often the top end of spend on retargeting is actually not incremental. We're almost always overspending and so we want to be in almost a scarcity mindset on retargeting and then an abundance mindset on cold targeting and new customer growth. Number two on retargeting is consolidation of audiences. Uh what is very like 2019 and you should not be doing is having a bunch of different adsets segmented by like

retargeting different product categories based on what landing page they landed on or retargeting add to cart separate from initiate checkout separate from website visitors. that hyper segmentation in retargeting. All it does is increase CPMs. It just makes it more expensive for you to serve retargeting ads and it has almost no direct upside. If someone can give me a really strong reason as to why their ad account has six different adsets targeting six different degrees of separation of warmth in the funnel, like I don't know what you're doing. The CPM increase is never outweighed by the

conversion rate increase. And then number three is really your core KPI other than obviously incremental revenue on retargeting campaigns should be frequency. You should just be looking at frequency very tightly to understand how many times are we retargeting existing customers, engaged audiences every single month, every single week and is this intuitively or subjectively too high or too low? Should we be retargeting existing customers 30 times a month, but most people probably not. And so intuitively, you know, you're probably overspent. So you can probably pull that down. Now, a big part of retargeting is DPAs, which

is dynamic product ads. Uh, DPAs are a pretty critical component of most ad accounts, particularly if we're talking about fashion, particularly if we're talking about large cataloges or SKs within a product portfolio. DPAs are great because they're going to dynamically retarget a user with a catalog ad that is going to rep prioritize products based on the products that they looked at on the website if everything's working correctly and if the pixel worked and actually fired the correct data back. Amazing. Now, the bad thing about DPAs is that they will pretty much only place right before

the moment of purchase as like a bottom offunnel most aware retargeting ad. And so, because of that, they get way more credit than they should actually get. The rorowaz on these things looks way better than it actually is. And so, people end up naively overspending on them substantially. And so, there's a few rules when it comes to DPAs. Number one is make sure that when you're assessing return on ad spend, you're assessing based on 7-day click or based on incremental attribution. If people don't know what I'm talking about with these attribution settings, just watch our

video that was posted on the 18th of May called if your ad account looks good but profit isn't, watch this video. It's an hour and 2 minutes long. It goes all into attribution models and data integrity so you can better understand that. Number two is that your DPA performance is really just a direct reflection of your topfunnel investment. If you increase budget on top of funnel, if you go harder on top of funnel, DPA rorowes goes up because you just capture more revenue at the bottom. If you decrease top of funnel, your DPA Performance will

go down. We actually made an entire video on this like 6 months ago, which was called the DPA death spiral of fashion brand. The whole idea was that I had seen six fashion brands in a row on six audits I did in the course of like 2 weeks where all of them were spending 80% of their spend on DPAs and the account was just declining over the past 6 months. And the reason being is that they saw the performance on DPAs. the agency just kept bumping budgets because it was good and then the brand as

a function of the agency's advice was like oh DPAs are doing so well do we need to make as much creative and they were like no you don't because DPAs are holding up performance so they just started decreasing creative production and creative and topfunnel spend allocation within the ad account and then the whole business just falls apart because if DPA sit at the bottom of the funnel and then you just stop spending on top obviously that is not sustainable then when we think about existing customer spend allocation what happened which was really bad for the

entire performance marketing and ecom space in my opinion was that Meta launched advantage plus campaigns like 2 years ago and the first setting that there was on advantage plus campaigns was this little box that you could tick and it was called existing customer spend cap and it was a percentage input. So you would input your percentage that you wanted going towards existing customer. This framed everyone into the mindset of not only thinking about existing customer spend, which was good, but it got them framed towards, yeah, what percent should we put towards existing customers? What should

our percentage be in terms of meta spend allocation? It's a terrible way to look at it. It's the wrong frame entirely because percentage is irrelevant. Who cares what the percent is towards existing customer? I only really care about two things. Number one, is it profitable? If we spend another $10 a day or $1,000 a day on existing Customers, do we make enough return revenue to be able to pay for that spend and make profit? Number two, which is kind of a softer metric that gives us more realtime visibility, is what is our frequency? Yes, sure,

we can have a 20% existing customer spend, but if that means that our frequency on existing customers is a 50 every 30 days, what are we doing? Why are we targeting people 50 times a month? Vice versa, if our spend is 20%, but we only have a frequency on existing customers of one, we're probably substantially underspending. And so your existing customer spend as a percentage is a bad metric because percentages change based on how big the existing customer pool is and how much budget allocation you currently have to new customer acquisition. Those two variables will

just substantially change the spend allocation as a percentage to where it's useless. So how do you think about it instead? You just think about it as a dollar value. So, how much dollar spend should we have per month to existing customers? And that you can calculate by taking your amount of customers, you times by your CPMs, and then you times by the amount of frequency that you want per month on those customers. And this tells you exactly what your monthly budget should be towards an existing customer audience. So, here's some settings that you should be

turning off and not using. And number one is flexible ads. I explained briefly why before, which is the fact that you can load a bunch of creative into one single ad, but the issue is that that ad performs well or if it doesn't perform well, we don't actually know what creative was driving performance or not. And so the issue here becomes that it doesn't follow the number one foundation that we had for account structure, which is data integrity, which allows for feedback loops. So we can read the data and then we can make decision. On

flexible ads, you can't read the data. So it's fundamentally a bad ad type. Now, the only reason why you would do this is to try to get around the page limit within the account. There's much better ways to get around the page limit. I already gave them to you. If you still can't do that, if you're going to do flexible ads, you need to put a bunch of creative that's very similar. And so that if it does perform well, we can kind of say that this type of ad does well because it's all similar. If

you put any kind of large variation of asset types under a flexible ad, you then no longer know what's actually working and you're just introducing stuff into the account that doesn't facilitate a decision-making feedback loop. Why we as an agency really don't like flexible ads on top of that is that you don't have control over cropping the images and so it can serve the one by ones and story placements. It'll randomly crop the story placements into feed placements. It's generally terrible for the actual quality of how the ad presents on the feed. What Meta can

also do sometimes is if you have a bunch of flexible ads, it will just string them together and create a carousel out of them, which also isn't ideal when a client hasn't approved a carousel. That's a string of random images that have been put in the account. So, generally not a fan. Next is cost caps or bid caps. I just really wouldn't be concerned with worrying about bidding strategies if you're spending under $200,000 a month in an ad account. It's just not the right bottleneck to be solving for within the account. The question that you

should ask yourself is, will the business double if we just really focus on bidding strategies within Meta? And the answer normally is no, because it's a low leverage opportunity. Now, is that to say that you shouldn't run cost cap? Not at all. Like if you know what you're doing, you can run cost caps on any spend level. More so, I'm saying focusing on changing and testing bidding strategies at lower spend levels generally isn't the biggest lever to be able to unlock growth. Number three is turn off one day view attribution or at least don't use

it within the decision-m of how you're flowing budgets at an adset level and what's working and what's not because it will substantially overattribute based on existing customer revenue. Number four is you want to check for all of the advantage plus creative optimizations. Um, particularly you want to go into your advertiser settings and go into creative sub dropown and then there'll be a bunch of automated rules that are turned on which will effectively flip your creative optimization settings back on over time if you don't turn it off at the ad account advertiser setting level. So, not

only do you need to do this on every single ad that you launch, which is incredibly annoying, but you also need to do it at the advertiser setting level. And then number five is you really want to make sure that your audience segments are set up correctly. What has happened a lot and that we keep an eye on is that the sync of the audience segment from Clavio to Meta will break periodically. And when that happens, your audience segment stops flowing through existing customers correctly and so you end up spending on existing customers thinking they're

new and obviously that starts to erode data integrity within the account and you start spending in places that you don't want to. When it comes to testing discipline in the account, when you're setting up any kind of test, you want to answer two questions that are very similar to what we talked about before, which is what's the stat threshold? So, How many conversions do we need to say that it's statistically significant? Is it five? Is it 10? What is that threshold? And then number two is how quickly do we want to actually learn? So, do

we want to spend to that statistical significance in a 3-day period, in a 7-day period, in a 30-day period? How are we spending to the threshold to then be able to move out of the test or say that the test was a su success and scale it? The way that you actually calculate it using these two questions is that your daily testing budget on a test is your target conversions times by your expected CPA. So what is your expected cost per acquisition divided by the test duration in days. For example, if our target conversions is

20, that's the threshold in which we'll call the test statistically significant. Our expected cost per acquisition is $100. We want the test to go for 14 days. Then the daily budget on this test is $143 per day. Now, this is also why the common feedback from Meta is wrong, which is that Meta says that you need 50 conversions every 7 days to be able to exit the learning phase and achieve statistical significance. The issue is is that when you try to back propagate this into daily test budgets, it just blows out and becomes ridiculous for

like 99% of avatars. So if you take this exact test case as an example, if we need to target 50 conversions and CPA is obviously the same and we need to do it in 7 days, that means daily budget here needs to be $714 per day. And that's just for one concept adset test, which is just ridiculous cuz if we want to put like five tests in or 10 tests in, suddenly you're spending $7,000 a day just on the testing campaign. If I was to compress this Whole training into just four questions that you should

write down and walk away from when you're looking to build an account structure is number one, does the structure produce data that I can read to make decisions? Because ultimately the whole point in the account structure is that we're producing readable data so that we can continue to iterate and make decisions within the business. Number two is does the structure match how the business actually makes? So is the structure commercially aligned to the objectives of the business and how the business is producing profit. Number three is the structure appropriate for the spend tier and creative

output. Creative throughput as well as the amount that you're spending is going to substantially change the structure that can be allowed within the account. And then lastly, are the support layers, so DPA, retargeting, website visitors, existing customers, is that aligned with the actual retention in the business? Do we need this amount of spend? How much spend do we need? How are we structuring DPAs and retargeting? How much spend distribution is going here? This is all going to be once again context dependent on the actual business and the retention within the business. If the answer to

any of these four questions is no, then the account structure is wrong and you need to restructure accordingly. So, that's the end of the video. Two things. Number one, if you're a performance marketer, we're always hiring. Please reach out to us at hiringblensedigital.com.au. If you're an e-commerce brand or retail business doing over $5 million a year in online revenue, click the link below. You can get a free audit from us where we'll break down how all of this strategy actually translates into your account. We'll give you the nuances of account structure. And if you're in

neither of those two buckets, subscribe. In 2026 and 2027, creative is the lever that sits within meta ads. Media buying fundamentally is not your advantage right now. You need to become incredibly good at creative strategy, at creative design, production, then moving that into the ad account and analyzing the creative to create this flywheel effect. By the end of this video, you'll know everything you need to know about creative for 2026. We're going to be running through why creative is the growth lever, how Meta's algorithm actually works in 2026, concept architecture, hook strategies that actually scale,

all the formats you can use and test, a testing framework to take that creative and actually put it into the account to get statistically relevant results. We're then going to move into creative volume, creative diversity, and a financial model that ties it all together. We're going to talk about fatigue, scaling winners, portfolio management, creative production. So, what did the teams, the briefing, and the process look like? And then lastly, we'll tie all of it together and package it into the 2026 playbook. So, with that being said, diving into section number one, we have why creative

is the growth lever. I've been media buying on Facebook since 2018. So, I've been about eight coming up on 9 years in the Meta Ads Manager. And I remember back in 2018, 2019, 2020, there was so much leverage in media buying. You actually didn't need to be very good at creative. And as long as you had a massive edge on media buying, you could beat a lot of people. And the reason for that is that there was a massive skill gap. In fact, I would argue that any arbitrage opportunity that exists within a market is

a product of the skill gap that exists at that point in time. So back then, the Difference between someone that just goes and opens up the ad manager and someone that is a super advanced media buyer that spends 12 hours a day doing this and they're three years in, there is such an enormous difference that that difference allows for arbitrage and allowed for the opportunity. Now over time, Meta has slowly compressed media buying. The targeting is now fully automated. I would recommend 99% of people are just running broadly, so there's no skill within interest targeting.

Andromeda now optimizes for the distribution of the creative. Consolidation is always favored across all of these digital platforms now because conversion data consolidates which allows for better machine learning modeling. And all of the media buying hacks that used to exist like duplicating adsets or using bidding models in a certain way. None of them really work that well anymore. None of them really give you a big arbitrage. And so because of that this arbitrage opportunity has compressed and instead where the big arbitrage opportunity exists right now is in creative. And it is because most people fortunately

for you are terrible at creative. And so because of that there is this huge opportunity that if you simply become good at creative strategy at understanding what a good ad looks like and then being able to build a system that can churn out a ton of highquality ads. This is where you now make all the money on meta. And by the end of this video, you should unlock this opportunity. So, the first thing I want you thinking through here is the portfolio model. So, let's say that you have $1,000 a day in ad spend and

there is two different scenarios here. Scenario one is you have one single creative that's holding the entire $1,000 per day and you're at a 4x row And then you have another pathway that you could choose. You've got three ads holding $1,000 a day together. And this is also at a 4x rorowaz blended across these three creators. Now, I would rather be in this situation every single day of the week. And the reason being is that this situation decreases risk enormously within the business because any of these three creative can fatigue and the other ones will

pick up the slack and you can continue operating at this efficiency, at this ad spend. But over here, if this creative fatigues, you're done. And I say this out of PTSD, working with probably 8 to 10 clients now over the course of the last 6 years who have come to us 40% down yearon-year in new customer acquisition, their business is in a terrible spot, and they're going, "Our business is declining. It's because we built this eight figureure business off three creatives that held a million dollars in ad spend and now these three creatives are all

fatiguing and we're trying to spin up variance and it's kind of keeping us at some reasonable level but we just can't get back to the level and we keep decline. And my response these days is it's too late. We're not taking on that account because you should have fixed that 6 to 8 months ago. What people will do is they will get one or two winning ads that can hold a tremendous amount of ad spend and then what they do is they just stop making more ads and they go like why would we spend money

on creative production? That's useless. Let's just double down and put all our ad spend here and then focus on other areas of the business. Yeah, amazing. Until these creatives fatigue and the only reality of creatives in meta ads is that every single creative will fatigue. Every Single creative has a maximum spend threshold that it will reach. And so you want to be constantly thinking through creative production, ad account management, and creative through the lens of this portfolio model, which is that we want to get as many ads as possible holding as much spend as possible.

We do not want to overlever into one or two ads because this is ultimately what ends up destroying a lot of businesses. Let me give you three common mistakes that most teams are making when it comes to creative. Number one is a lot of people confuse activity for strategy. I'll give you an example of this, which is recently in the last few months, I've seen two ad accounts that are launching 2,000 plus new ads a month, and they're barely able to crack a4 million in ad spend per month. Now, for those that work on large

ad accounts, you would know that that is insane. You do not need 2 to 3,000 new ads a month on a4 million budget. That's so much volume. And they weren't profitable. And so you're looking at the fact that they're launching so much volume, they're not profitable, they're doing so much activity, but there's clearly a gap in strategy because if you were putting 2,000 great creative into the account with a good offer, with a good persona, with a good angle, then you shouldn't need anywhere near 2,000 ads a month. And so, yes, I think volume's I

think you should be launching as many ads as humanly possible, but with a standard of quality and with a strategy behind the creative that is actually going into the account. So, do not mistake activity for strategy. This is really common right now because people think they can just launch more ads and their brand will scale. In fact, I hear this all the time, which is people are like, "Okay, I need to double my business. Do I just launch 40 more ads a month?" I was like, "If it's that Simple, everyone would do it." There's a

little bit more to just increasing activity. The second one is misdiagnosing bottlenecks. So with increative, there is a ton of different bottlenecks that exist that could be holding you back. It could be the quality of the creative. It could be the format diversification. It could be the quantity that's sitting at the different stages of awareness or the stages of the funnel. So you have nowhere near enough top offunnel assets to produce the scale that you actually want. This might be a bottleneck on not actually introducing partnership ads or employing different strategies into the ad account

to unlock more spend. There's so many different bottlenecks that exist just within creative that holds a business back and most people don't understand all the bottlenecks that exist. So they misdiagnose it. They don't realize what the actual bottleneck is. And so then they focus on activity that doesn't actually move the needle. For example, those two brands that I spoke about before, they were not only mistaking activity for strategy, but they were also misdiagnosing the bottleneck in the business, which is that they thought just increasing sheer volume would suddenly unlock growth, but it's not doing anything

for them. And it's because they're trying to fix something that actually doesn't need to be fixed. And then the last one, which I hope people are getting over by now, but I still see it from time to time, which is that creative is seen as a cost center rather than a revenue driver. And this is fundamentally because of the shift away from media buying to creative, which is that back in the day when media buying had a lot of arbitrage behind it, you didn't need a lot of creative and a Lot of volume of creative.

And so because of that, business owners would often see the ad spend in the platform, the spend on Google, the spend on Meta, the spend on Tik Tok, Pinterest, Microsoft as revenue driving, but the actual creative that's used to fuel that spend as a cost center. We want to minimize that cost as much as possible because we just want the platforms to find us customers. But that's the complete wrong way around. The creative is actually what is driving the revenue and Meta and these other platforms are just driving the distribution. Now, the fact that you

can actually just post this stuff on organic and get free distribution as well means that creative is also the revenue driver organically. And so, if you're thinking that creative is a cost center, you've got the completely wrong approach. We're going to be diving into later in this video exactly how much spend you should have on creative production respective to your ad spend level. So, you can know if you're currently under or overinvesting in creative. Before diving into creative strategy, you first need to understand how Meta's algorithm works because this is the distribution tool for the

assets that you're putting in it. Now, everyone's probably heard the word Andromeda. It's a massive buzz word. I've talked about it in a lot of content. And to put it simply, it's just an algorithm change that Meta made because they had a lot of GPUs for training AI for their metaverse that kind of fell flat. And so, they repurposed all of those training chips towards a better ad serving platform. with that the mechanical shift that you need to be across is that before andrometer and to be honest we weren't even doing this pre-andrometer but it's

an easy way to simplify the concept and then I'll go a little bit more nuanced which is that pre-andrometer you would choose the audience so you would say hey Here is our interest please target this interest and you would then go and load all the creatives in and so you put creative one creative two maybe you're putting three creatives under this ad set and you're saying here's the interest here's the ads go and match them. Now, post Andromeda, what happened is you actually do the opposite. So, you put the ads into the account and you

don't select any targeting and then Meta goes and looks at the creatives and it goes, "Okay, I think that these are going to resonate with these people based on the transcript, based on the words in the ads, based on the copy, based on my understanding of historical conversion data." And then it will go and find the people to target. Now, this is fundamentally a way better system because interest targeting has a lot of downside. Okay? When you're doing interest targeting, you're effectively labeling people with an interest, let's say pets, and then just arbitrarily serving all

your creative to people that are interested in pets. But the reality is is that it's not just people interested in pets that want your product. It's probably people that are interested in pets within a specific age range that also have other variables or psychographic data points that are similar between them. And so there's so much additional nuance that needs to be captured and this serving mechanism ends up capturing that nuance better than the old one. Now the reality is is that meta was kind of already doing this pre Andromeda. We haven't used interest targeting in

most accounts for like 2 years. But what Andromeda really did is popularize this new approach to structuring an ad account. And so there were still so many agencies and so many people running interest targeting it was Crazy. Even though we have a video that was put out two years ago about this exact topic on the meta algorithm and this was before Andromeda. So I was explaining this concept before this whole update even rolled out and why we don't use interest targeting. It's called how machine learning works in 2025 on meta ads. But this has popularized

this approach and now this is really how you want to be thinking through the asset. Now why this becomes important is because the creative that you loads in that you load in is the targeting. So, if you load a bunch of creative that's all the same, you end up just targeting the same person. You don't reach new people. If you end up loading out a bunch of creative that isn't clearly resonating with a specific target demographic, then it's going to get confused. For example, if you're calling out everyone in your ads, if you're like, "Hey,

everyone, look at this product." Meta is going to be very confused cuz it's going to go, "What? Should we target the entire population? Is there a subset? Who should we actually go after?" You also need to think through this targeting mechanism at an adset level in terms of structuring because if you just put a bunch of ads together that are all targeting different people under the one adset, the adset will also get confused. Who are we targeting? So there's a lot of additional nuance that we'll break down here when we get into the structure and

testing section of this video. But this is fundamentally the key concept that you need to understand here. There's also the creative similarity score that rolled out in meta recently. Now, most people still can't access this in their ad accounts, but if you reach out to your dedicated meta rep, they can get you this score. So, you can have an Idea of what your similarity score is in the business. And as it currently works, the higher the number, the worse it is because the more similar your creatives. Now, why this matters so much is because of

Andromeda bundling. And so, if you have, as I said before, multiple different creatives here, let's say 1 2 3, but they're all slight variants. Let's say they're the same campaign shoot, but the model's just standing in a slightly different position. Or let's say it's UGC, but maybe there's a slight variation in the script halfway through the video. Or let's say these are actually very different videos in terms of formatting, but they're all speaking to the exact same audience. Well, what ends up happening is Andromeda will bundle them together and serve them all to the same

pool of audience. Now, because of this, because they're similar creatives going to the same audience, what happens in the ad account is frequency goes up. So frequency is how many times users are seeing your ads. Obviously, these ads are just going to serve to the same people over and over again. So frequency goes up. Your reach will go down. So you'll stop reaching net new users. And as a function of this, over time, your return on ad spend will go down. And so you'll start seeing diminishing returns in the platform. So instead, the position that

you actually want to be in here is launching three creatives into the account, but these three creatives talk to different people and therefore reach their own audiences. And with that frequency declines because you're reaching different people with these ads, it allows you to unlock more spend because as these scale, you can almost think about this circle expanding As it goes to colder and colder audiences. But because these are talking to different people, you can see there's not much audience overlap. And as a function of that, you're going to be able to scale to new audiences

effectively without just increasing frequency and seeing efficiency decline. Another incredibly important concept here about how the platform algorithm actually works is in sequencing of creative and the reliability of return on ad spend as you decline through the account structure. So when it comes to sequencing creative, the reality is and I think everyone watching this will agree with this which is that if you just serve one ad to a user once, will they buy? And the answer 99.99% of the time is no. Generally speaking, you need to see an ad more than once to be confident

in purchasing. Now, yes, there are some people out there that are getting one hit by ads and suddenly buying. All right, that does happen. In fact, it's probably happened to me at one point in my life, but it's very, very rare and you definitely cannot build an efficient business off of those people. And so, because of that, you need to serve them generally more than one ad. Now, you could serve them this ad again a second time. However, Meta put out a article ages ago. It was like four or five years ago which said that

after two impressions of an identical creative probability of purchase declines precipitously. And so if you look at the probability of someone buying, let's say this is a percentage and then you look at the amount of ads that they are seeing. So this is frequency of an identical ad. So we're talking about the same ad here. You're not showing them different ads, just the same ad. When you show them the ad once, There might be a certain percentage uh chance of them buying. Then when you show them the ad twice, it might be the same. It

actually might even be slightly higher because on second viewing now they're warmer and they're more likely to buy. But then on third viewing it declines massively and then on fourth massively and then it pretty much declines to zero. And so there is very little purpose at all in showing a user the same identical ad more than twice. And the way to actually stress test this idea that I'm telling you right now in your own ad account is you can pull your ad account up. You can go down to the ad level, look at the last

30 days, 60 days, and look at frequency at an ad level on cold campaigns. And what you will see is that cold ads almost never have a frequency above a two. And if they do, you have a creative fatigue issue. You do not have enough diversification and volume in your account. And that is a red flag. any good account that we're running, which is all of them, um, no one has a frequency above a two on a cold out just doesn't happen because if we see this immediate red flag, we need to fix it. We

need more diversity. We need more volume. Now, the big kicker here is that what meta are published in that statement, and if I can go and find that article, I'll put it in the description below, is that if you then go and introduce a novel creative, the probability of purchase stays high. And so after seeing the same ad once and then twice, if we go and serve a different ad in as the third impression, probability stays up here. And then if we go and serve a different ad as the Fourth impression, probability stays up here.

And so we can actually keep probability of purchase up by rotating in new novel creative. Now the question becomes, how does the algorithm do this? And how does it decide the sequencing of creatives to these users to try to get them to convert? And this is where sequencing comes into the ad account structure. When meta is optimizing across serving users ads, they're optimizing across understanding what sequence of creative actually gets the purchase. So is showing users ad one then add two then add three going to get the click and purchase or is it showing users

ad three as the opening creative then add two then add one going to get the purchase or is it some other kind of combination? Are we coming in here and then going back here? There's all of these different combinations of serving these ads in a particular order that might yield a click and might yield a conversion. Now, the kicker here is that meta only attributes based on last click attribution. So, if we assume that the first sequencing example was what is actually occurring, they're going 1 2 3. If the click occurs here, all of the

return on ad spend, all of the revenue gets attributed to this ad and it looks really good. Let's say it's at a 6x row. Add ad two. Maybe some people buy at this stage in the sequence, but not a lot. And two at a 2x. And then on this top ad, we're down at like a 1.4x. When you look at this and you look at spend distribution, let's say spend is equal across all three creative. Naturally, you go in here and you go, "Oh, this is Obvious. Turn ad one, turn ad two off, cuz ad

three is getting all the performance." But that would be a bad idea. And this is the concept in a lot of our content about do not turn a creative off when you are hitting KPI. If this is hitting your performance goal overall as an adset, don't touch anything because Meta is sequencing between these creatives on purpose and it is creating an outcome. Meta's optimization is not based on last click. It's really important concept to understand. Meta's optimization is actually based on multiclick and view. And so Meta sees that people are viewing this ad, viewing this

ad, then clicking on this ad and buying. And therefore, it distributes spend here knowing that it played a role in the purchasing journey, but you don't see that. Unfortunately, we don't get any visibility into this at all within the ads manager, which is actually really annoying. And so, as a product of that, we just need to trust the system. We also need to think about the creatives that we're putting into the account. How are they working through this purchasing journey? Normally, if you're good at creative, if you're good at creative strategy, you can look at

these three creatives and you can go, "Oh, it's pretty obvious to me that this is actually the opening ad. This is then probably the second ad." And then this is the converting ad because this is a super unaware topunnel creative. This is then a product aware creative. And then this is a bottom of funnel very aware creative. And so it's super obvious to me just looking at the three ads that that's what the sequence looks like. But if you don't understand creative strategy, you're going to look at this and go, I don't understand what's going

on. Why are the rorowes like this? Turn off, turn off, turn off, destroy performance. We're going to go into this in a lot more detail later on once we start going through the testing structure and actually KPIing and Understand what is a good ad, what is a bad ad. But this is a really important concept to be across. Second concept to be across is that the reliability of rorowaz declines as you go down in the account structure. So everyone who's opened up a meta ad account knows the basic structure which is you have campaigns you

have adets then you have ads right the campaign you set up the budget typically and you set up the overall optimization event so is it optimizing for sales reach etc ads set this is where all the targeting occurs now people forget that these days because people go oh targeting is just broad so I don't even really know what's happening at the adset level but it's really important to reinforce targeting is occurring at the ads level and then down here the ads This is obviously the creative that gets served to the end user. Now, when we

look down at the ad level and we start assessing creative, the issue here is everything I just went through, we can look at the rorowaz numbers, but they're kind of meaningless because their last click attribution and there's a multi-click or multi- view through journey occurring here. And so, I can go in and say, "Yeah, this has a 6x row as amazing. Let's like try to scale it or make more ads like that." But it's often a bad decision. So instead when we move up to the ads set level right here now we can look at

all of these returns as a group and we can go let's not go into the individual ads because the reliability of uh the attribution here is terrible because there's so much happening across these different sequences. So let's just go up to the ad set and look okay at the adset level what's our rorowaz here and maybe overall it's at a 3x. Now this is a lot more reliable. It is worth noting that cross serving of ads still occurs at the Adset level. So you can have uh multiple different ad sets. So if I just throw

another ads set in up the top here, we'll call this ads set two and let's just put a couple creatives under here. A sequence that can occur is that people see this ad and adset one. They then see this ad and then they get served another adset. Now it's not prioritized by the algorithm. The algorithm will always prioritize serving adjacent ads under an adset due to the way that targeting is siloed. But there is cross adset targeting. And so because of that, we might have a 3x here. We might have a 4x up here. This

adset seems to be doing better. But it's actually just because this adset is picking up some bottom of funnel conversions being generated from this adset. That can be completely avoided. As long as your structure is good and you're diversifying concepts across the adset level, you shouldn't have much cross targeting. And you can measure all this and we'll go into it later on. But the whole idea is adset is better. We can trust this number. we can trust it more than the ad level. Then once you go all the way up to the campaign level, the

campaign rorowaz number, as long as it's on 7-day click attribution is super reliable. This is generally very congruent with the actual P&L, as long as it's on 7-day click and as long as existing customers are excluded. And so because of that, when we're making decisions within meta, when we're thinking about structure, when we're thinking about introducing creative, we want to be thinking about measuring at the adset or the campaign level, not the ad level. And we want to be thinking through this idea of sequencing. So the practical implications of Andrometer and the way that the

algorithm and ad platform Currently works is four-fold. Number one, you need concept diversity. We're obviously going to go into what a concept is, how to break it down, how to come up with concepts, what diversity looks like later on. But number one is you need diversity within the creative going into the account. You al also obviously need that introduced into the structure so that you're having a concept per adset so you don't have any of this cross targeting going on. Number two is you need format variety. Okay? You can't unfortunately just have the same format

of all of your creative. And the reason being is that meta will prioritize formats that people resonate with. I'll give you an example which is that I never ever and I could scroll my Instagram feed for the next 3 hours and show you. I will never get a piece of UGC under my profile. Just doesn't happen. And it's because I never click on them. I never resonate with them. And so Meta inherently knows that lowfi UGC content should not be served to me. And so if your ad account is just loi UGC, I'll never see

it. Now, same thing actually for DPA ads. Now DPA are very bottom of funnel. This isn't really cold. I'd have to be already very very product aware on the website. But I don't think I have ever been served a DPA almost in the history of uh the ads that I get served on my personal profile. Now that is likely a product of the fact that DPAs are not a creative type that I resonate with. I never click on them. And so if you just had UGC and DPAs, it's very unlikely you'd ever reach me. So

you need to be making sure you have as much diversity in format type. And I'm going to be giving you all the formats and how you should be mix and matching them later on so that you can reach as many people as possible. We then have hook quality. So this is going To be a big section of this video, but with Andromeda, any kind of persona call out that occurs within the actual creative is going to go and prioritize that persona. So if I make an ad and at the start I say, if you're a

business owner doing over $5 million a year, Andromeda is going to read the transcript. It auto transcribes every single ad and it's going to go, "Okay, let's try to find a similar audience to what he just said. What if my ad or my content has nothing to do with that audience? Well, then we've just push the creative to an audience that's not actually relevant. And people do this a lot. People's hook actually isn't as good as they think it is. People call out a persona without problem agitating. There's there's so many different issues that actually

lie in hooks, which is why we're spending so much time on it cuz I think people are terrible at hooks. Most ads I see, they're just not good. And then lastly is that with Andromeda, volume is absolutely critical. The spend capacity of most ads isn't very high. Most ads that you enter the into the account, if we look at mean ad spend per ad unit, sits at about $700 to $1.5,000 depending on the ad account. And so generally, you can't get much spend through most ads. So you need volume to compensate, but you also need

strategy at the same time because if you're just uh putting sheer volume into the account, which I've seen so many people do with no actual strategy behind the volume, it doesn't actually work. So let's run through concept architecture. This is how you need to be thinking through the segmentation of different creative. This is how your naming conventions should be set up on meta as well. This is how you should be ideulating in a Google sheet or whatever format you want to think through this. And we'll also talk about the AI implications into this ideation process.

So what actually is a concept? A concept is a persona, angle, and offer. And so to have diversity within the creative that's going into the account, you need one of these three variables to change. If the offer changes, it will resonate with a different audience, with a different pool of people because it's either a different product or if it's it's a different value proposition or packaging of the product. If the angle changes, obviously, it's going to resonate with with someone different because you're problem agitating on something else. And so, if you sold skin care, you

might start by problem agitating on wrinkles, but then you might have another ad that problem agitates on that. They're going to reach two very different audiences even though it's the same product because the angle changed. And then lastly, we have a persona change. Now, the easiest way to think through a persona change is obviously who are we talking to, but also who is actually in the ad. And so, if there's a 60-year-old woman in the ad, it's obviously going to resonate with a very different target demographic than if there is a 20-year-old woman in the

ad. And so, because of that, we can adjust the persona and that will inherently adjust who resonates with the creative and who it will target on the platform. And so one of the easiest ways to ideulate here and come up with almost infinite creative concepts is to start ideulating on who are all the different personas that we want to reach. What are the core personas in the business? And you can go through and you can probably come up with two or three or four. And then if you really drill deeper, you can come up with

tons more. So yeah, we could go like females age between 40 to 45 etc. But then we can start digging into like what are the actual pain points? And so females aged 20 to 25 dealing with acne. Females aged between 30 to 35 dealing with acne which is a slightly different problem because acne shouldn't be occurring at that age. So maybe it's hormonal acne. And then we can go women aged between 40 to 45 with acne as well. That's also a completely Different problem that should be fixed by now. And so what are the different

underlying causes that we can then resonate with in that persona? And so you can build personas out in ridiculous detail based on all of the problems that the product solves. By the way, this is the case in every industry as well as fashion. So, in fashion, people really struggle to get their head around concept architecture, but with personas, it's the same thing. Most large retail fashion brands will break their target demographic down into like four core personas, and they'll give them names. But the reality is, you can go way deeper than this because a persona

could be a particular problem that a particular person is facing. And so there's this particular person who's a mom aged between 30 to 35 with young kids. And the problem that they face is they need an affordable dress that can go from a work dress into a night dress on a Friday afternoon. That is a persona. That's a persona that you could make infinite creative for for that exact particular painoint and circumstance. And you could probably think of 400 examples of that. When we move into angle, it's the presentation of how we're actually uh going

after the problem. So what is the angle? What is the value proposition? Why would you purchase the product? And then the offer is obviously really straightforward, which is what are we actually offering? The key here is personas aren't demographics. They're archetypes that are defined by problems, desires, and buying triggers. So, we want to be thinking about who are our avatars. You want to think hard on this. Spend 20 minutes on it. Then, what problems do we actually solve for these people? Then what are the differences in pain points, language as to how we talk to

this Person and the proof that is required for this person. And once we understand these three, we should be able to build out a whole lot of personas which you can then once that is done manually go to an LLM like Claude or ChatBT or Gemini. Give it the entire persona list and tell it come up with 20 to 30 more and give it as much context as possible which you should already have loaded in there in some capacity about the business about the target demographic etc. Claude, OpenAI, LLMs, they're only going to give you

averages. And so if you just give it three or four personas and you give it your website and you say, "Hey, come up with some more personas," you're going to get an average outcome. And everyone's going to be doing it and you're going to get the same as them. And you're going to revert to the main. So what you want to do is if you are going to do this, you need to put all the manual effort into actually building out proper 30 to 40 personas that are actually good. And then you need to give

it an unbelievably wide amount of context as to the business and the target demographic and everything you can so that it can actually idiate properly on another 20. And of the 20 that it produces, you should just be putting that in there. You should be refining down. Probably cut down to four to five. Give that back in. Say these four to five were good. Ideate on another 10 to 20. And then that way you can beef this out from 20 to 30 personas or whatever you get to and double the number. One really important note

here is that people die in persona work by keeping the personas too broad. broad personas end up producing forgettable creative that is nowhere near specific enough to the target demographic that you want to resonate with. And so if you come away with a persona while you're building this out, which is, let's say, women age between 25 to 45, this is a terrible persona. The creative that You're going to make for this audience is nowhere near specific enough that it's going to resonate with enough people to be able to hold a good amount of spend. as

an example, instead of going for the persona busy professional, which I actually see so much when when I tell people to do this concept buildout, and instead you would want to delineate this into a consultant who flies twice a week and needs a carry-on wardrobe that works for boardrooms and bars. Way more specific. And now with that persona call out, you're already thinking about how to talk to this person. You can almost visualize who this is, what their problem is, what we need to solve, what the creative should look like. It gives so much more

specificity into the actual briefing and production process. And then guess what? thought we actually reach this audience properly. But when we say busy professional, what is a busy professional? That is so wide. There are so like the age is all over the place. Their gender is all over the place. What do they actually do? What do they get up to? Why do they even need this product? This is so broad that you will you will end up with mediocre creative. You are better off making five ads for this target demographic than five for this. Now,

five for this will scale harder if you can actually do a proper unaware ad that is going to resonate with this entire pool of people. Yes, it's a bigger total addressable market. So, yes, you'll be able to hold more spend. But the reality of you making a good ad for an audience this broad is actually very unlikely. And so, you are way better off making more ads for very specific personas. And then as you start to become a eight mid 8 figure brand, you can start to play around in trying To get ads to work

for very broad audiences. But the skill level required here is so high and it is just not required for most ad accounts. most ad accounts you can just go hyper specific and get way better results than needing to go and try to target the entire of a population. Another example here is instead of doing, let's say, healthconscious parent, you could target a mom who's tried six different supplements because her toddler won't eat vegetables. Once again, way more specific. Now, you could go away and you could write a script for that. You know exactly what that

ad looks like. You know, who should be in it. You know who you're talking to. You know what the proof needs to be for that particular target demographic. you know what the language needs to be and you know the pain points of that particular persona. Now an angle is the specific argument or perspective that you take when you're speaking to a persona. This is where a lot of creative strategy actually ends up failing because brands default to product features rather than crafting genuine angles that resonate with the persona. So I'm going to give you a

lot of tactical application here. Number one is you want to problem agitate in some capacity. You don't need to do this in all of your creatives. In fact, especially if people aren't even problem aware, then you can't even problem agitate, right? But you want to be leading with pain as much as possible. And you want to make the viewer feel the pain, feel like they have been heard before you go and introduce the solution. Another great angle here is approaching with a contrarian truth. So you want to challenge a common belief that people have which

allows them to reframe. An example of this, particularly just as a hook, is everything you've been told about X is wrong. Everything you've been told about the protein industry is wrong. And then You go into obviously an educational piece that then might problem agitate that then obviously loops into a solution. You want to typically always have social proofing in the angle. You want to build some kind of authority or expertise in the domain. So this is more specifically domain expertise. So you don't need for example a doctor and a lab coach showing up in every

single ads for every industry. Right? If you're selling a running product, you want a marathon runner showing up. You want someone with authority in the domain that you're trying to sell in. Curiosity is great and is commonly in most angles that do well. And this is specifically creating what's called a curiosity gap. And so when you open up a video asset, and you can actually do this in image assets, too, but when you open up the video, you open up and create a curiosity gap or an information gap. As an example, you'd say, "Here's what

the top 1% of millionaires know about saving money that they don't want you to know about." And then you don't answer it straight away. And you go into a story and you go into something else. You can go into any of these other aspects. You problem agitate, you have a contrarian truth, you go into some degree of social proofing, although that's probably stacked towards the back of the ad. And then eventually the ad goes on and on and on. You pull people through this journey, you educate them on everything, and then you end up actually

answering the curiosity gap. And so you didn't have any information. You want to get to the information and eventually the information is provided to you. And so that's how you hold people through an ad. All good angles also have some degree of comparison or objection handling. And so often people who are watching your ad have tried to solve their problem through other solutions. And so you want to tell them why your solution will work versus all of the competitors. You've probably tried this. You probably tried this. You probably tried this. If it's a piece of

user generated content, you'll commonly see the actual person in the script going, "I tried this just like you would have and it didn't work for me. I then tried this. I thought it was overhyped. It didn't work. And then finally, I was like, okay, I'm going to give one more thing a shot." And it actually worked. And then this one is the least compliant, but it is the best performing component to add into an angle. So, I'm going to throw it in, which is a transformation, aka a before verse after. Uh, before verse afters, Meta

does not like them. You have to sneak these into creatives in a way that the algorithm doesn't flag it. But these are some of the best performing creatives ever because people want to see visually a before and after. I think this is basic human psychology. Before and afters will always crush for as long as they exist. And because of that, Meta doesn't like you running them. So, when we're thinking about the angle, we're starting at the persona and then we're thinking about what angle and approach do we need to be able to resonate with this

target persona. And what you will notice is I've given you a lot of tactical application that can be included or built together into a script that creates the angle. So, the real key with the angle is this. You can simplify it into a one sentence, which is if our persona is women with kids aged 35 to 40 that are having energy issues, we're going to then sell them creatine powder and the angle is going to be that it solves energy issues based on particular studies and then we can continue chaining it out. The thing is

the angle realistically if you're going to write this out properly, we're just talking about the script. Okay? And so you just Want to write the whole script. If it's a static, this is obviously going to be a bit tighter and you could probably encapsulate this in a few sentences. But the idea is we want to figure out what angle are we approaching selling this person on with a myriad of different strategies that's going to resonate with them. And so if we take for example that woman age 40 to 45 with creatine and energy issues, well,

we're going to want social proofing, but we want to we're going to want social proofing for that particular persona. So we're going to want either case studies around persons people like that or the clinical studies that we're going to reference which is going to give us expertise and authority needs to reference their particular age and gender. We want to think about problem agitation on that particular persona. So how are we going to agitate to that problem? Are we going to pull in some kind of contrarian truth or curiosity gap that's going to hold them into

the asset? Can we do any kind of comparisons to stuff that they've tried in the past? Has this persona tried things or is this very new to them? And then can we pull in any kind of transformation at the end as a piece of social proof to back into the creative? And so we choose the elements that we want. We think about how we're going to slice and dice these into a script and that becomes the angle. And we're going to build out some actual examples of this in a moment. But before we do, I

want to talk about the testing priority order. Which of these three components has the largest impact? What should you be thinking about testing the most of or the least of? Number one at the top is angles. So this is exactly what we're just discussing. This is the message or the argument. The reason why this is such a high priority is because we can sell to any persona that is relevant to the product. You should be able to Effectively look at who has bought from you in the past. What persona do they sit in? Why did

they buy from you? And then we just need to craft an angle to that particular persona. If that persona isn't purchasing from us, we know that it's actually really an angle or an offer issue. It's not an issue with the persona cuz we've already proven that these people buy from us. Okay? And so when something's not working, we typically don't want to look at the persona as long as you did it well. Like let's not do broad 25 to 45year-old females. Let's get hyper specific into a very specific person that has purchased from us before

and let's target them. But then ultimately it's going to come down to an angle or offer issue. Which is why angle is the first thing you should test. If things aren't working, you need to rotate out and improve the angle. And then the second is the offer. Now fundamentally the offer actually might be the number one most important thing out of all of this. But the issue is is that you actually don't have much flexibility in changing the offer. What do I mean by that? The offer is two things. The offer is the product and

then the offer is the value proposition or the way that we're framing value. Now, when it comes to the product, if you have a single product and ads aren't working well, I can't really tell you to change the product. Right? Now, you could you could change the whole business, but the reality is is that the product is the product, we need to change the angle or the persona, right? So even though this is the most important thing, the product ultimately is king out of anything, we don't have much flexibility here. Now if you have a

very large skew count and maybe this is a new product that's being launched into an existing 20 product portfolio, then sure, then we can go and say, is this an offer issue? Is this an issue with the product? And we can start to troubleshoot accordingly. But for most people, you're pretty stuck on the product. Then when it comes to the value proposition, the value proposition, We've got a couple options, right? We can go and try testing bundles. We can go and try testing like a gift with purchase. We can go and try all of these

different repackaging of the offer, right? We can do like tid as well, so you get percentage off when you buy two or three. We can go and throw in like a subscription offer. So, there's a lot of testing that we can do here. And we do this with a lot of clients is we'll come up with a bunch of different offers and how we should reframe value to try and get the funnel working more effectively. But ultimately, you are limited here. There's only so many offers you can test. And we run into this issue which

is like we've tested like seven variations of how we can spin these two to three products and now we're relatively stuck and the only way that we can increase the value perception on the offer is to just do harsh deep discount and we don't want to do that. We don't want to go and just eat 40% into margin and build a brand off a heavy discount reliance. And so although the offer is unbelievably important, this is a huge variable that we do a lot of testing on and we do a lot of offer testing for

clients. It's something that eventually you're going to hit a wall on. Like there's only so much offer testing you can actually do and once you've found the best performing offer, you're stuck on it. And so that is where then we want to be doing a lot of switch ups in personas. And personas is ultimately what's actually going to give us breadth and volume in the account. And then I have number four in here. There is actually a fourth component of concept design. Now this didn't really used to exist pre-andromeda, but it's an additional component that

we've added in recently. And it's because of the way that Andromeda will take different formats and serve it to different people. So, as I said before, I don't Get UGC content. I don't get dynamic product ads. And so, you could do all the mix and matching that you want of personas, of angles, of offers to try to hit me. But, if the only ads in the account is UGC and DPAs and so there is a fourth component to be able to reach more people, which is the format. You can have the same persona, same angle,

same offer, and you can change the format and you will reach a different audience. However, this is nowhere near as effective as these other components. If you do this and then you change the format, the reality is you're probably going to hit a lot of the same people cuz you are talking to the same people. You are going with the same angle and you are offering them them the same thing. You'll get a bit of novel reach. You'll obviously reach people that don't resonate with the current formats, but this isn't like a huge unlock great

variable in the account. And that's why it is down at number four in the testing priority. Now, the big issue here and the reason why I spend so much time going through the testing priority order is everyone gets this completely the wrong way round. People spend all of their time testing formats. They go, "Oh, let's test some UGC cuz UGC will work." UGC as a concept doesn't work. It is the script. It is the persona. It is the angle. It is the offer that is contained within this type of content that is what works. Just

like a professional photo shoot doesn't work. The format isn't what works. It's what's contained within the format. Then people move up to the persona and they go we need to reach different people. Okay, we understand that Andromeda we need to talk to different people. So let's talk to different people. This is the second priority. It is not because if your ads aren't working fundamentally you have bad messaging and you have bad arguments for why they should buy. The creative is just bad. And creative being bad is normally a function of the angle and the Scripting.

And so this is what you should be fixing. Changing up the persona is just you trying to target different people with terrible ads. Then they go to the offer and they say, "Well, all the big influencers, Hormosi says offer is king. That's the thing that matters the most. We need to go to the offer." Sure, but you're going to once again fatigue offer testing pretty quickly. There's only so much you can do with your one or two product. And then they end up at the angle, which is their lowest priority, but it should be the

highest. So, this is how you need to be thinking through testing priority. Now the practical application of thinking through this format type actually translates directly into the account structure which is that when you have a campaign you want your adsets to be segmented out based on concept. And the reason being is that the targeting exists at the adset level. And so if you go and put in as I said before a bunch of different ads that are a bunch of different concepts resonating with different people well then the ad set that's trying to figure out

who to target is going to get confused. It's going to go, "Wait a second. This is resonating with men, old men. This is resonating with young women. Like, why are there all these different angles in here? Who do we actually target?" Now, yes, there is some siloing of target demographic targeting at an individual ad level, but a lot of it still sits at the adset level. And so, you're going to get worse performance when you go and bundle all different concepts together. Not only that, because you can't trust the reliability of rorowaz at an ad

level very well. When you go and put all these together and you don't hit KPI, you do hit KPI. Let's say maybe you got a 3x rorowaz here. Well, then you're going to look at these ads and not actually understand what concept works because there's three concepts here. So, what is the learning from this adset? Well, if it was all under one concept, one persona, one angle, one offer, we could go, "This concept works really well. Let's go and make more like this. Let's double down in this direction. Let's not overlever. Let's think about portfolio

management, but we should definitely put a little bit more resourcing here cuz there's a squeeze available." If instead you have three different concepts in here and it's hitting KPI, what what's the conclusion? is is all the concepts working, but this one doesn't have a good rorowaz. But we know that the rorowaz at the ad level isn't reliable. So like what is it? How do we draw conclusions out of a campaign structure with mixed up concepts at the adset level? You can't. It's very difficult to do. You just got to hope your ads are good and

that you get a good return. But if you want to actually have any kind of learning feedback loop, you want to be structuring the account based on concepts. All right. So, here's a full buildout that I've just put together of a persona angle and offer that I think would actually crush the formats we can add as we go along here. So, and you're going to have to excuse my uh writing. Persona, casual runner training for a sub 4-hour marathon. It's a male. They're a gym goer. So, they're not just a runner, but they've gone to

the gym. They follow XY Z influencers. So, there's particular influences that we've called out, and we know that they're probably following them in some capacity. and they make about 70k a year perom. This is important in terms of framing value around what we're actually going to try to sell to them. So where do they actually sit in terms of income level? In terms of the angle, we're going to come in with science backing because we know that this person probably follows someone like Andrew Hubman. So we can come in with at least an angle around

him. We're going to go Through how creatine improves recovery, but it doesn't improve recovery in general because we want to talk to this persona. So it improves recovery for the long run. Now people are only going to be doing a long run if they're training for a marathon. That enables a fiveinut improvement on the marathon time. So we're directly connecting the entire angle to the persona and we're giving them an expected outcome. Then the offer is relatively straightforward. Creatine gummies buy two get one free. Now this offer was selected by myself just due to the

income level here. So, we probably want increased perceived value cuz creatine gummies are generally quite expensive and it's going to be hard to push it on to someone and to reframe them against creatine for running the actual format here. This is where we can chop and change this and we can have like 10 different ads, right? We could go into a VSSL format. Probably not going to work as well on this age demographic, but it is a potential option. We could go into UGC, which I think is going to be great. We could go into

founder talking head. We could go into a more educational hi-fi styled piece. Uh you could also actually translate this into a carousel. Formats go on and on, but they're probably the starting ones that I'd go for. VSSL I might just leave out for now because the age is too low for a VSSL to work on. And so that is one concept where now we can come into here and we can start changing one of these variables to be able to mix and match it to different target demographics. And so the reality is that this angle

will also work for women that are training for a marathon. So we can change over that persona and we have now a whole different uh concept that we can start making ads for. Now we can tweak the angle a little bit and Rather than making it for long days, we could make it for people that are trying to get up to a 5k run. And if you're just started running and you're trying to get to a 5k run, creatine is a great way to build up initial improvements in recovery and therefore we can now resonate

with a different persona. So you can start to mix and match this accordingly to reach way more people. Now on hook strategy, everyone thinks that the hook of an ad is a trick to try to get someone to watch the video. But the hook of an ad is a play towards relevancy. What you were doing is promising the user that what they're about to watch is relevant and contextual to them. And that distinction matters enormously because a lot of people think that they should just be baiting people into watching. They should be trying to hook

them through some kind of meme at the start. And yes, that'll give you really high hook rates if that's what you want to optimize for, but it's not actually going to give you conversions. Now, there's actually three different layers to the hook. You have the visual. This is what's actually on the screen. You then have audio. This is obviously the underlying audio. And then you actually have the copy that overlays the visual. This could also be the primary text that sits above the ad. Now, the reason why I'm even putting this as a section in

the video is because the hook is so important in video assets because 80% of people never watch past the hook. And so, if you're going and serving ads into the marketplace and 80% of people never even watch your ad cuz they don't like the hook, so they just skip straight past it, well, it's pretty critical then for us to try to optimize the hook. Okay? If most people are never going to see past this point in the video, then let's make sure this video is relevant, providing value in some capacity or trying to pull people

through the video that are relevant and need to watch. An example of this in terms of resource allocation Is that you can spend an hour going and making two hooks and then five different bodies for this ad. Or you could go and spend the same amount of time but just make two bodies and then 15 hooks. And you will probably always get better performance here given that the actual scripting is good on the bodies. And the reason being is that the delta available in performance increase is normally higher in the hook than it is in

the body. Once again, assuming scripting is good. If you're not good at scripting, well then like one of these bodies might actually be a good script and then it will hit and you'll get a misrepresentation of the actual impact here of the different components of the ad. Now, in terms of rating whether a hook is good, cuz I do this a lot. I do a lot of reaction videos on Instagram saying this is a bad hook, this is a good hook, and I'm rating how well people have done in putting a hook together. Well, there

is actually a grading system that you can think through as to how good your hook actually is. Number one is clarity. Can a stranger understand what you are actually talking about within the first 3 seconds or are you just trying to confuse them as a way to hook them in? Number two is relevance. Now, people mistake this for a persona call out. They mistake this for me taking the persona that I was talking about before and just calling them out at the start. Hey, if you're a mom that has kids and you're age 30 to

35 and you have this problem, well, listen up. Persona call outs are super overrated and it's because no one wants to hear themselves called out. People want to hear the problem that they're actually dealing with. And so you'll always see much Better performance agitating on a problem than you will on any kind of persona call out. And I have this from personal experience. Anytime I get an ad saying agency owners who need XY Z, listen up. I never watch the ad. I move on. But if they agitate on a specific problem, it feels much more

genuine. actually wants to I actually want to watch through and find out what the solution is to the problem that I'm actually currently facing. So you want it to be relevant, but persona call outs are nowhere near as effective as just making the problem agitation relevant to the person. We then have novelty. This isn't a requirement, but this is a really good way to create white space within the ad that you're making. And so this concept of whites space is also this concept of purple ocean theory, which is that you have a blue ocean over

here. You then have a red ocean over here. Now, this is actually one of my favorite books that I've ever read. I actually reference it in a lot of content, blue ocean theory. This is that you want to reposition your product into a market that differentiates you from everyone else so that you're not competing on pricing. You're not even getting compared and you're seen as a completely novel product or service within the market. Red ocean means that you're selling the same stuff as everyone else and then you're just getting priced down to zero. These are

terrible businesses. These are incredible businesses. The issue with blue oceans is they don't really exist because if you're going truly blue, you're selling something that's never been sold before and that's kind of risky because there's no actual market knowledge. You need to do tons of education and there's no proven product market fit either. So yeah, you could reposition into something that you think people might want, but they might not even want it. And so you realistically want to sit like somewhere in the middle here where you're repositioning it into something novel, but it still has

proven product market fit. It's still proven that it is going to work. It's just not what everyone else is doing. it's Slightly adjacent. And so we do this a lot in our own content in terms of format testing. So we'll rotate in different formats because we want to test new things. We'll take ideas from other industries, from other people. We'll come up with our own ideas to try to cut through and have novelty that people haven't seen before. I would say this YouTube video is actually an example of that, which is that there isn't many

creative courses on YouTube for meta ads that are 2 plus hours long. So you want to be thinking through that concept too in the hook. Now, once again, it's not a requirement, but it will help in standing out without having some kind of gimmicky tactic to hold people through the ad that's not actually unique. The question I would be asking yourself here is, has this hook been done a million times? And if the hook's been done a million times, probably don't do it. The next is specificity. So, is there a way, and you're not going

to be able to do this in a lot of hooks, but in our specific hook, is there a way we can use numbers, maybe quantify names, or maybe even quantify outcomes right out of the gates, that's going to build authority into the hook. Anytime you can start a hook by name dropping someone that's super famous, that's relevant to the product in some capacity, it always does better. If you can name drop some kind of number, 67% of people got this result, then once again, it's going to build authority. If you can speak straight to specific

outcomes, it's going to build authority and it's going to be more specific to your target demographic than you want to resonate with. And then lastly, we have credibility, which is fairly obvious. Now, with credibility, you're usually not going to be able to do this through the copy or through the audio. It's typically going to have to be done through the visual. And so, you are Going to need either like someone famous that pops up at the start. You're going to need someone that is uh contextually relevant. So, if you're selling supplements, you have a doctor

who pops up. If you're selling running supplements, it's obviously like a runner that pops up. You want to be thinking about how you can build credibility. Now, you don't need 10 out of 10 scores on all of these to be able to have a good hook. You really only need a couple of them done well. This is how you should be thinking through hook creation. Now, let me give you a bunch of hooks that you can go and start using. Number one is problem agitation. So, you want to lead with the pain that the viewer

is already feeling and you want to make them feel seen. An example of this is, are you struggling with back pain from desk work? This will only resonate with problem aware audiences. So you can get really good performance, but it isn't going to be scalable above that stage of awareness. Number two is a contrarian truth. So you challenge a common assumption. You create cognitive friction that demands a resolution. An example here is everything you've been told about protein timing is wrong. And this works across all levels of awareness. And it works strong and organic and

paid. Number three is specific proof. So you want to lead with a measurable outcome, numbers, time frames, specificity. An example of this is I lost 12 kilos in 90 days without giving up pasta. So you build credibility throughout the hook. The next one is a curiosity gap. We spoke about this before which is that you create an information gap that can only be closed by watching. An example is what the top 1% of brands know about meta ads that you don't. And a funny side note is that my average view time when I use that

hook or some variation Of that hook is over 45 seconds. And so that hook actually works incredibly well because it creates an information gap in my own organic content. Truth bomb. There's actually one specific client where this is done unbelievably well for. So that's why I threw it in here because I don't see many people doing it, which is that you lead with an uncomfortable truth about the product's price, ingredients, or industry. The honesty itself becomes the hook. And so the hook in this instance is something along the lines of this product costs $120. Yeah,

it's super expensive, but it's our best seller and everyone buys it. Here's why. It ends up working exceptionally well for premium or higher average order value products where the price ends up being the main objection that people have. So you confront it way up front and it actually works really well at cutting through on cold audiences cuz even if I've never seen the product before, if someone's opening with this thing is expensive, I'm like, "Okay, what is that thing? Teach me more." And then it can go into a very very educational piece. Obviously that's only

going to be relevant to very specific industries. The next is psychological confrontation. And so this is being direct and patent interrupting with a statement that feels like a friend calling you out. An example here would be, you know, that drawer full of products you never use, that's the problem. Or stop buying things that make you feel guilty for not using them. It's not clickbait, but it's playing on an emotional pain point, and so the viewer ends up feeling personally addressed, which ends up acting as a scroll stopper. Once again, all of these things also apply

into organic, which is where I think getting good at organic content is actually a very critical skill for getting good at creative strategy because it just translates into paid. We'll keep going here, which is you can also do sensory I'm going to call this a texture hook as well. This does unbelievably well in fashion right now. This is just an ASMR hook. So, you don't need any words. 2 to 3 seconds close-up of the fabric being handled, a product being squeezed, liquid being poured, package being opened, something ASMR related. These do incredibly well. Uh we

have a lot of ads right now that are high performers. On this hook, you've then got a founders letter. You can go and apply all of these into most brands. So you want to open with the founder and their origin story. So we started this because dot dot dot or here is why I left my career to go and build this. The more risk you can build within the hook as well, the better. And so you want to build tension. And this is once again a big organic strategy that you just want to translate into

paid, which is that you want to have high stakes within the hook. if you're going to have a founders letter style hook so that people have a reason to continue watching. I reorggaged my house and put $300,000 down to be able to order my first order of this product and here's what's happened since. Like that hook I could almost guarantee would do well. Now obviously that has to be a real circumstance which is why that hook does well, right? It's things that you can't fake. If you actually did put so much risk on the line,

it's a great angle to be able to talk about that risk. The last one here is social proof. So you want to lead with some kind of credibility marker. 90% of customers saw results in 30 days. Now this is nine different hook types that you can go and take. You can probably apply at least seven of these into your particular brand right now. There are infinite hook Strategies, right? I could write hooks here all day long. I could fill the whole board with 50 different hook types. You could go to Claude or JBT right now

and say, "Give me all the different hook types that I can run." The idea is that you want to take hook strategies and you want to translate it into your brand, but the scripting is going to be critical. And where you're going to really see uh a standout from competitors is over here in this novelty piece on whitespace. So you can become an 8 figure brand by using these hook strategies and having relatively good scores over here. But if you want an ad that's going to hold half a million dollars in spend, you do probably

want to generate something novel at the beginning of the creative that's going to stand out from competitors. And then what's going to happen is all your competitors are just going to go and copy you and so you will set the trend. Now I've spoken about the hook a lot but an an important subcomponent of the hook is the bridge which is where you move between specific components within the video. Now the two big bridges that often occur is when you go from hook to body and then when you go from body to CTA. These two

bridges often don't end up being done well. And so what ends up happening is if you think about product awareness on the y- axis and then you think about time going through the video asset on the x- axis, what ends up happening is people will hook you in and you'll become aware that it's an ad and then suddenly as you bridge to the body, some people will just instantly introduce the product. So they'll hook you with something like, are you trying to train for a marathon and you want better recovery? Well, creatine from Blue Sense

Digital is your solution. That ad won't do well. Hook was good. bridge ruin the whole thing. The whole body could be good. Like the body scripting could be perfect, hook scripting could be perfect, call to action could be great, but the bridge Ruins the whole asset. And so you don't want your asset to look like this where there's this massive step change in product awareness as a user goes through the creative. Instead, you want to blend these out so that people are slowly becoming aware that they're being solved to. And so instead, you're looking like

this. And so I don't really even need to give you that much information on how to do this. To me, it's pretty intuitive, which is just blend in the introduction of the product. Don't just suddenly introduce it. Or else people go quickly from, oh, I was watching something that was educational or that was teaching me to, oh, I'm actually being sold a product. And people don't intuitively like that feeling. This also, this jump in bridge will typically ruin the flow of scripting in most very good top offunnel unaware ads. And so when it comes to

really good topfunnel unaware ads, what they do is they move people through the stages of awareness as you progress through the ad. This is actually something that VS Cells video salelet letter ads do unbelievably well because they get the capacity to teach through the asset. And so you start with unaware. So you're hooking people in with something that they might be interested in, usually some kind of curiosity gap. Then you're making them problem aware. Okay, now I understand that there's a problem with drinking water. If I just drink water every day, not only does it

not have electrolytes, which apparently I need. I didn't even realize this was a problem, but also tap water is incredibly bad for me. Okay, that's interesting. What's the solution? Oh, the solution is that I need this amount of salt in this kind of ratio, and I need to stop using tap water. Okay. Oh, there's actually a product that solves this problem. It's canned mineral water with electrolytes in it. Oh, okay. I was unaware of anything. I became problem aware. I became solution aware. I'm now product aware. Okay, this is going to fix all of these

problems that I didn't realize I had. And now I move down into the aware stage of awareness where maybe I convert on a retargeting ad. Now, this is generally how you want to be thinking through scripting on really good topunnel ads. But what most people do is that they'll start really unaware. They'll then go problem aware and then like in here they'll just introduce the product. You're like, what are we doing? We haven't even moved them to the next stage of awareness. We can't introduce the product yet. And people do this cuz they think in

retention curves. And so they go, well, if we look at the retention curve on an ad, which is how many people are still watching on average as we move through a creative, the retention curve will look like this. And so intuitively, you go, well, 80% of people are going to drop off by the 90 secondond mark. So we need to introduce our product before the 90cond mark cuz we don't want most of the users to not know they're being sold our product. It's the wrong way to think through it, okay? You actually want to introduce

your product as late as possible in the creative because you want people to be problem aware and s solution aware and really preframe and engaged ready to buy once the product is actually introduced. You don't want to introduce the product too early before they're actually aware of what the problem and solution is. And so you often end up getting way better conversion rates out of an asset by moving the product introduction and call to action way later because these people are now way more qualified to actually buy and are way more higher intent. And so,

yes, you can introduce it sooner, but they're not high intent enough that you will actually have worse conversion rates. And even though it's more total People, you will ultimately get worse total purchase volume out of the asset. And so, that is something to be really careful of when you're thinking about bridging. Now, to wrap this section up, one final question that I get a lot is, does hook testing still work post Andromeda? Because isn't the whole point in Andromeda that we need diversity? And if I have a 92 Facebook ad, me just changing the first

3 seconds at the start, that's not diversity, right? That's the same ad. We just changed the first 3 seconds. Meta will still recognize a different hook as a different creative. It will get its own creative ID. It will not get pulled together with this asset. It will go and reach a novel unique audience. Now obviously there is going to be some degree of overlap because 87 seconds of this asset is going to be talking to the same person with the same problem with the same offer. So if we actually look at the concept framework it

is the same concept. It's just the hook change. So the question becomes is hook changes worth it? Yes. Why is it worth it? Because it is so low effort. Rotating in new hooks is almost free. Okay. When you record the ad, you just record the hooks six times and it takes an extra few minutes and now you've got six different hook variations. Or you can take existing high performers and you throw new hooks on them. We have had ads that have spent $50, $100,000 fatigued. Then we just rotate 50 new hooks on and we get

another $100,000 to spend at the same efficiency. And those hooks took like one hour to shoot. And so you can very much so keep winners alive through changing hooks. You can also take ads That didn't work and then suddenly make them work by fixing the hook if it had a really bad hook rate. So yes, absolutely hook testing still works post Andromeda. You should be doing it. It is still the highest leverage point in videos, organic ads, YouTube, everything. All right. So when it comes to formats, there's five questions that you need to be asking

yourself when you're thinking on a format. Number one, what is the concept? We need to define the concept first before we even think about formats. You want to go out build your persona angle offer then come back to thinking about the format. You don't start with the format and work backwards. This is the last step. Secondly, does this concept require education? Now, if it requires education, you likely need to go into some longer form asset that's going to educate. Number three, can the value prop be communicated in a single frame? This is obviously going to

lean yourself towards images, carousels, something that's easier to produce, can obviously still be in a video, but you do have an advantage with images that they're very easy and cheap to produce and you can produce a lot of them. So, anytime you can produce statics over videos, you generally want to do it. Now, there's two different approaches to statics if we just leave these out for the moment, which is number one, you can use statics to quickly test different angles and different concepts, get fast feedback, and then develop that into a video format where you

can go a lot deeper and create a more scalable asset. And then number two is that images are very cheap to make, and so they often end up outweighing videos in the cost of production. So if you look at the average spend of an image versus a video in your account, it might actually make a lot of sense to just go way harder on images because the cost of production is super low and the average revenue Generation is decent in proportion. And so this is what is called cost of production against the average revenue per

ad. This is something that we calculate for everyone during the order process because we want to understand what is the average revenue per ad that you put into the account. For most people, it's about like $2,000ish. And then what is your cost of production against this delineated down to contribution margin? And so we minus off gross profit. We minus off the cost of serving the ad. We might find out that average contribution margin per ad unit is, let's say, $350. How much is it costing us to make an ad? It might be costing us $400.

Okay, this equation doesn't work. We're currently paying more to produce an ad than we contribution margin out of each ad. We can then go and delineate this down to images specifically. We might found out images cost us about $50, but they're producing us $300. Videos are costing us $500 and we're only getting $400 here. And so we actually find out even though videos do perform better, the actual cost of production versus expected contribution margin is worse than image assets. And therefore, we should actually just double down on images. And I know people doing uh $200,000

a day in revenue on just AI images. And so there's absolutely no reason as to why you can't do hundreds of thousands of dollars a day on images only. And if you can't, it just ends up being a skill issue. Number four is, does the product need a demo? This is just going to be a general question that applies to the business as a whole, which is how much education do we need to force people through? Because obviously you're going to be relatively restricted with images then unless you get really fancy and you have a

good designer. And then lastly, which ties into both of these points, is what stage of awareness are we sitting In? Okay, if we're sitting in a very high stage of awareness, we need a creative that is going to be a longer format that can educate, typically a video. If we are sitting at a very low stage of awareness, like the product aware or aware stages, uh we can just go for DPAs and images and get away with very bottom ofunnel ads. I want to reiterate that the belief that image ads are bottom of funnel and

video ads are top of funnel is fundamentally a skill issue. Once again, I have seen multiple ad accounts. I've talked to multiple people that do hundreds of thousands of dollars per day in revenue off just static. And so if you're saying, "Oh, statics don't work. They're bottom of funnel assets." It's because, yeah, the way that you design them and the way that you approach them and concept them out is bottom of funnel. But you can make top of funnel statics. We're now going to cut out to three ad breakdowns that I have done on static

images. We're going to go through an organic tweet format, which is an IM8 health ad. We're going to go through a creatine gummies ad. And then we're going to go through a GLP1 hair loss ad. And you will see three statics that are holding tons of spend on top of funnel. This is an incredible static ad. 4.7,000 likes down here. It's by IM8. And let's break down exactly why it's so good. Number one, it's a new format. They're actually using a tweet here and they're reposting it so it feels organic to the platform, but it's

actually an ad. Now, further to this, they're reinforcing it by using not a partnership ad. So they're not collaborating with IMA and Dr. James, but it's just Dr. James. So, they're running it as effectively a whitelisting ad through this profile. So, you don't know that this is an ad. In fact, you would look at this for a very long time, and it's only until you start getting through the copy that sits under the creative over here that you actually Realize that you're getting sold something. So, what's so good about the static over here other than

it being a common format and that you not really knowing that it's an ad, so it feels organic? Well, symptoms of vitamin deficiencies. So right away we're problem agitating into okay here's some education around how vitamin deficiencies could be impacting you then this is an unbelievably wide set of symptoms. So if you have dry eyes, if you have poor vision, if you have low energy, if you have bad skin, if you have cracks on your mouth, if you have fatigue, if you have bleeding gums, if you have painful muscles, if you have excessive bruising, all

of these things, very, very wide, total addressable market, well then it's an issue with a vitamin deficiency, and it tells you exactly what the vitamin deficiency is. So you get education around, okay, here's my problem, here's the solution. Then you start probably writing a copy and you go the best multivitamin equals IMA daily ultimate 10% off. Here's the thing. Then you get more educational pieces here about how the product actually works. And obviously because this is an ad, a big call to action will appear at the bottom here that it will allow you to click

straight off the site. So a problem agitates to a very wide target demographic. It then educates about exactly what the solution is in terms of vitamin to solve your particular problem. It does it in a format that's super organic. It then maintains that organic feeling by not having the brand and company name attached with the post. And then this is obviously driving off to cold audiences, probably driving off to this guy's audience, which is even better because we're obviously piggybacking off his organic. This is an avitorial static ad. These are really good. They perform incredibly

well. They only perform well though if you go to an educationbased landing page or you use the primary text to educate heavily. So, let's break it down. The best hair growth products of 2025 and the ones to avoid. Here's what the big brands are not telling you. And then we have a bunch of different products here. So, we know or we at least assume, okay, this is some kind of blog. This is some kind of article that's going to educate us around what hair growth products we should actually be buying. Now, even better, it's running

through a doctor's page. Okay, so there's a bit of authority here. As a dermatologist, I set out on a mission to find the most effective hair growth products. Not very long copy, so they're not selling us on anything here. Okay, so we want to find out more. So, what do we do? We go and we go and click learn more. Now, from here, we end up on a website called Dermatologist Reviews. Now, this website is owned by the brand that's trying to sell you on something. This is a really good funnel. So, hi, who is

the person? Why do they have authority? This is the blog posted. I set out on a mission. And here are just the products ranked. And then it goes into all of the reasons as to why. And it goes into selling that product. And then it does go into the other ones, but it goes into why, yeah, I probably wouldn't choose these. They're not the best. Now, why is this a giveaway that this is actually a brand pushing? Well, because if we look at the rankings here, you can't click on any of these, right? But you

can click on this one. And if you click on this one, goes to a custom landing page that then takes you through a quizunnel. You go through a quizunnel and then you get pushed through to the conversion. So, this is a really good funnel that they're running here. I really like it. And almost to make sure they're a little bit compliant, I imagine. As you scroll down, this takes You to the same funnel, but as we continue going down, look at all the green ticks here. So many green ticks, one red. Only available online. And

then him and hers. Oh, widely available online. Oh, that's the best thing about it. All of these negatives, but you can still click here and it takes you to an Amazon affiliate link. And so, they're like, h, if people are not clicking on us, then we may as well still redirect them somewhere. And then obviously, you can see they put the effort into reviewing all of these and all the reasons as to why they wouldn't buy them. This is a super effective funnel. I imagine this is doing quite well for them. If your target demographic

is 50 plus, you can serve on Facebook feeds. This ad type crushes. It's static image ads with a long form copy. And the reason why they do so well is they go through a massive story that speaks to an unaware audience and slowly pulls them into being problem aware, pulls them into being solution aware, then pulls them into being product aware, and then finally they're all the way at the stage in which they can convert. These ads will crush on cold audiences, but it takes an enormous amount of skill to actually be able to write

this copy properly. Like yes, you can be AI assisted, but if you don't know fundamentals of how to emotionally appeal to a specific target demographic, then it's not going to work. And this is ultimately where getting good at copywriting really matters. If you can get good at copywriting and you can get good at pulling people through the stages of customer awareness, then that is ultimately how you can create these static image ads with long form copy that are going to absolutely crush. Now, in addition to that, to keep it contextual to 2026 and the time

in which this video is being made, a type of video asset that does incredibly well right now on older target demographics is a VSSL, a video sales letter. The reason why these types of ads do so well is because they pull people through all of the stages of awareness and they're Highly educational. You can also really leverage AI within them. We make a lot of just AI VSSLs end to end and they do quite well. And then some of them do unbelievably well and hold hundreds of thousands of dollars in ad spend. So what we're

going to do now is I'm going to pop up a few examples of VSSLs and me breaking them down so you can see an example of what this ad type looks like. You have IBS, leaky gut, or bloating? You're probably missing this one thing. You're This is a really good VSSL. First off, it starts with a persona call out. If you have these issues, keep watching. Doctor probably told you to just manage it with a course of pills. That's because then it's going into objection handling. The common way to fix it. They're treating the symptom,

not the cause. But here, then a problem agitates and opens a curiosity gap. Okay. I thought it was actually going to cause us to fix the problem, but this ad is saying it's not. All right, I'm going to keep watching. Here's what they're not telling you. Your gut isn't broken, it's starving. Your gut is an eco. I don't love that because it's an AI script. So that's called contrast negation, which is where you say it's not X, it's Y. You see a lot of this in a lot of scripting these days. I probably would take

that out. An ecosystem of trillions of bacteria. When the good bacteria die and bad bacteria take over, everything collapses. That's when you get inflammation, a leaky gut, and your body stops absorbing nutrients. This really good educational piece. This is dispiosis, and it's not rare. 40% of people have it. Most of them don't even know. You might be among them. But here's the problem with really good way to enforce that. No, you do Have this problem. Because when you prescribe a problem to someone, particularly when it's like a disease or it's a physical medical issue, the

immediate reaction is, well, I probably don't have this. I imagine this is rare. So, instantly increasing the tamb by 40% of people actually have this problem is a really good way to keep people watching to go, okay, maybe I do have the problem. Tell me a little bit more about it. With every gut supplement you've tried. They just throw probiotics at it. That's like sending a single soldier into a war zone. One probiotic can't survive in a toxic gut. It dies. It doesn't colonize. It doesn't help. IM8 Daily Ultimate Essentials doesn't just add probiotics. First,

digestive enzymes clear. I don't love the bridge. If the ad team is watching this, I would extend for another 15 seconds and educate around how the IM8 ingredients actually fix the issue and then introduce IM8 as the solution that has those ingredients and undigested food. Second, prebiotics create a safe environment for good bacteria. Third, 10 billion CFU of probiotics flood in and colonize. Fourth, postbiotics strengthen your gut lining and stop the leaks. So the 10 to 15 seconds would be saying uh the way that you get probiotics to work is that you have prebiotics and

postbiotics around it and that's actually what's creates the system to work. Fortunately, IM8 has put all of those three solutions in one sachet. So you can get them all at once and it can fix uh the issues that we agitated at the start. It's a complete military operation, not a single soldier. That's an AI script once again. Contrast Negation. I try to avoid that. And it works in a 12week clinical trial. 85% of And then it goes into social proofing. It keeps going on. Overall, really good ad. Just some small points of improvement. This ad

has 25 million views. It's been running for 3 to 4 years now. It's absolutely crushing. And I'll show you what it does so well. It takes a really common format of VSSLs and then it adds objection handling UGC to the front. So, let's click play and start to watch. I want to talk to you about something that's been making waves on social media lately. You probably seen this guy all over your feed claiming that cutting carbs is so they're using the split screen for engagement and to add more context to the video. Is it necessary

for weight loss? No need for long hours on the treadmill either. Now, I know it sounds too good to be true, but let me tell you, there's more to this than meets the eye. I decided when you're watching this, you think that this guy is obviously going to disagree with the cutting carbs because he's coming from a news authorative perspective. And it kind of feels like a call out video. Like, this guy is talking about how you don't need to eat carbs. It's crazy. Like, what is he talking about? I decided to delve into this

topic and do some research. Turns out this guy is a celebrity trainer who's gained a massive following for challenging mainstream advice. He claims that what we've been told about weight loss is wrong. Completely wrong. And you know what makes a lot of sense when you listen to what he has to say. So that's the first time in which he validates the person he's talking about. But everything through that first 28 seconds was contrarian hooks. Contrarian takes over and over again. Celebrities don't cut Carbs. That ain't how you lose weight. I don't do that. None of

my clients do that. Nobody that has ever asked me how to lose weight have I ever told them to do that because it simply just doesn't work. For anyone who doesn't know their metabolic type, listen up. The fastest way to burn fat and get shredded. It's not keto. It's not paleo. It's not carnivore. It's not vegan. Note that this creative is just going very very hard on contrarian takes. That's pretty much the entire strategy of this whole ad is like how do we put as many contrarian takes in here as possible to continue to string

people along as we kind of educate and we kind of hint towards that we have our own system and then eventually we obviously bridge into that system. So they're creating a curiosity gap through this whole thing. Okay, if no one needs to cut carbs then what is the actual solution? And it's definitely not super intense exercises like this because listen, if you're trying to get in shape, and I don't care what it's for, a wedding, vacation, summer, don't care. If you're trying to get in shape fast, you need to follow a plan for your body

type. And there is a very simple breakdown of body types. There's three body types. Uh you have the skinny person who's trying to put on muscle. Then you have someone who's kind of in the middle, maybe has a little bit too much fat. Turn's going to go into the educational piece. is effectively going to sell his system and the way that he views weight loss and body types and programming, Etc. This is such a good ad. In the last 5 days, it's got 2,000 likes. Let's break down exactly why. She hop on chess is opens

with a contrarian hook and obviously the visuals pull you in. What I would say if I was a caveman that wasn't aware of all the way builds contrast and then goes into an educational pace that you can boost it naturally. Being low tea can be caused by many. switches over to a different format which is the faceless format which is better inducive to education particularly on organic and we're trying to make this ad feel organic. High stress, vitamin deficiencies, not enough fat in your diet, not enough good sleep, microlastics, high estrogen, high inflammation. It's become

a serious problem for young men who often feel like there's only one solution after. So notice they'll problem agitate and then they'll educate. Problem agitate then educate. So there's all the problems. Here's the solution. This isn't the solution for you. goes back into problems. After all, global testosterone levels have dropped over 30% in the last 50 years. Young, fit, healthy. So, there's the education and then here's the problem agitation. Guys, we'll have the testosterone of a 70-year-old that has never seen the inside of a gym in their life. It's a grim side effect of our

modern society. With low tests taking away the feeling of being a real man, a consequence of this is a wave of young men, even teenagers, hopping on TRT out of desperation and because they don't want to fall behind. But injecting test is a minefield of side effects. Rapid hair loss, high estrogen causing gyno mood Swing, objection handling around the core premise of the hook, which was injecting test. And so now he's explaining why that's not a good idea. Infertility and acne, not to mention a big ass needle injected multiple times a week. In amongst this

chaos, how do I, a lifetime natural, have more testosterone than most guys running cycles? Whilst I'm doled with my sleep, training, recovery, and diet, so are many guys who are low tea. However, my secret weapon is a Natty Plus Ultimate. And there you go. There's the product. So, the product's been introduced after about a minute and a half into the creative, which is why it's done so well. It's why it's scaling so well. It's on an influencer profile. People know this influencer. It's likely prioritizing getting served to his own audience. And so, when people are

watching this, they're not knowing it's an ad. They're thinking it's educational. And honestly, it's a great educational piece. It goes into everything you need to know about testosterone and why you would want to get behind and buy this product. And it does it in a way that's very entertaining. It's fast cut. It's constantly changing formats. It's built native for the platform, which is why it's scaling so well. One important PSA here is DPAs. Okay, DPAs, dynamic product ads. Those are the carousel ads that will dynamically pull in the product that someone visited on the website

as long as your pixel is set up correctly, pushing back in data on which pages were viewed. DPAs cause a death spiral in a lot of brands. I see this in fashion particularly, uh, which is that DPAs have really good attributed rorowaz. the rorowaz number is very high. Now the rorowaz actually isn't very high once you delineate out existing customers. So if you remove existing customers it drops precipitously. And then you also want to delineate the attribution model down to Incremental attribution. And so you can do this by clicking on the breakdown button in the

top actually sorry you click on the columns button compare attribution settings incremental attribution apply and then you will get a split out of what the incremental rorowaz is on the ad. A quick lesson on how incremental attribution works is that every single day meta is targeting let's say 10,000 people in this square. It will hold out 10% of users and it will look at these people that didn't see your ads. How many of them converted? Let's say 1% of them converted. Over here, the people that do see your ads, 2% of them convert. Okay, great.

Or we can do what's called in data science is a difference in difference calculation, which is simply minusing the control versus the treatment. This equals 1%. So your ads are actually having a 1% impact, not a 2%. And so when we see a five rorowaz on that DPA, let's say as an example, we actually need to adjust that down by 50% to 2.5. So that's what incremental attribution is doing. That's what it's going to show you on the rorowaz here. and the rorowaz on your DPA is going to be way lower than you think it

is, particularly on new audiences. I am still yet to see an ad account where this is a great scalable ad type on cold audiences. Now, is it an ad type that can work on cold audiences? For sure. Can you spend $1,000 a day on it? For sure. We have some ad accounts that do. Can you spend $50,000 a day on it? Absolutely not. Like, you just can't scale DPAs that well on cold audiences. Most people are way overspending because the rorowaz number looks good. to just be super careful cuz what ends up happening there is

because DPAs sit at the bottom of the funnel down here is that people start reallocating their budgets from these topfunnel efforts that have bad return on ad spends and they just start moving budget down here. And when you start moving budget down here, you start losing the quantity of users that are coming through from the top of funnel and you start scaling something that isn't scalable and the whole funnel topples over and return on ad spend declines and everything goes poorly. So in fact when you're thinking about funnel budget allocation you want to be thinking

about 70 to 90% of budget going towards top of funnel you want 20%ish of budget going to middle of funnel and then you want really like 5 to 10% of budget going to bottom of funnel. There's actually a better way to calculate bottom of funnel budgeting which I'm going to show you later in the video. But this is rough split on how it should look. Now what's interesting these days is that you used to do this through campaign structure. So you would have your top ofunnel ads and you would just set your budget here. You'd

have your middle of funnel. You'd set your budget here and then you'd have your bottom bottom of funnel. Uh the issue right now is there actually is a lack of campaign structure. You shouldn't be structuring like this. You should just consolidate these down and you might be able to have a bottom of funnel that sits separate. And so because of that, the actual budget allocation is coming from the creative production volume. And so you want 70 to 90% of the volume of assets you make to be at the high stages of awareness. And you want

20% of them to be at the middle to bottom of funnel stages of awareness. So this is more so a percentage allocation of creative production than it is a percentage allocation of budgets across a campaign structure. I would say one of the most underutilized formats in 2026 is partnership ads. Now I think people are Catching on to this actually and the arbitrage is probably coming towards its end. So by the time you watch this video if you're way into the future this might not be the case anymore. You might have missed it. But partnership ads

what they are is they run through the creators profile. So you'll have your profile here. to your company name on the top of the post and then you'll have and and then the influencers profile right here. Now, these days there's actually a setting that allows them to change the combinations and so sometimes you'll just see the company, sometimes you'll just see the influencer, sometimes you'll see both and it will split test and decide on whatever the best mix is. But the reason why these ads do so well is that the targeting uses the audience on

both handles. And so if this is an influencer that's actually somewhat contextually relevant to your target demographic, which I hope it is, this ad will go and reach their audience and it will use their audience data to find new people. And so you're effectively leveraging the influencers's audience to help in your targeting on cold, not even just targeting a warm. And that's why they end up doing so well. They also feel super authentic. When you get an ad from a uh influencer profile like this, it doesn't feel as much so like an ad. So it

doesn't feel like you're getting sold to. So when you're getting an organic piece of content, it resonates better. In some accounts that we're working on, we have ad spend up to 40% of total account spend running through partnership or collaborator ads. And so this is a really important thing that you should have in your account. If you are not running them, you were just at a disadvantage to competitors. All right, so going into the testing framework and how this then translates into the ad account. This is what the testing cycle looks like, which is that

you build a hypothesis, which is you define what you're actually testing. So when we're introducing new creative, generally the Test is the concept and particularly one of those four variables that we're changing in the con. Then we want to build a bunch of ads under that concept. One big mistake that a lot of people make is that they come up with a concept, they build it all out, it looks amazing, and then they make one ad or they make two ads or they make three ads. Like really, if you want to test a concept properly,

you want to at least 10 plus ads considering that most people's hit rates in their account are around about 5%. And so technically speaking, if you have a new concept coming in, you would need to do if you had a 5% hit rate, you'd need to do at least 20 ads to expect a winner. And so thinking about your hit rate in the account as a function of how much volume you need to do to be able to validate the hypothesis is critical. Then you want to go and launch and I'm going to show you

with the exact structure that you should launch into. You then want to go and analyze particularly against the hypothesis. So what did we expect? What how much spend did we want to go through? How many conversions did we expect? What was the ROI? And then how did that compare? If it worked, we want to be looking at how we can employ strategies to scale. And then we want to go down and iterate. And this iteration process gets faster and faster the bigger your ad spend. So now when we think about account structure and introducing tests

into the account, this is where there becomes a lot of flexibility. And this is where we prefer not to actually make broad sweeping account structure recommendations because the fundamentals of account structure is that the complexity of the ad account should be a product of the complexity of the business. And so if you run a startup that's doing $10,000 a month in revenue and you have one product and you have one persona that you're trying to crack, the account structure should be unbelievably simple. You don't need a 100 different campaigns. On the other hand, if you're

an $800 million fashion retailer with 50 new SKs coming out per week and you have all of these different business units and regions within the business, well, the ad account's probably going to be pretty complex because it needs to match that complexity in some degree so that we can align the commercial goals of the business with the structure that is being run in the ad account. And so the structure I'm about to give you is a very broad sweeping this will work if you're a small to midsize business, but it's probably not exactly what you

should run right now or moving into the future as it will likely evolve as the business dynamics evolve. But to keep it simple, you generally want a testing campaign. Now, whether this is an or a CBO is honestly going to come down to personal preference, and I'll tell you how to think about this problem in a moment. But then underlying this, you have your concepts at the adset level. So this is concept one, this is concept two, this is concept three, etc., etc. Now, you're either setting budgets across these manually using an or the CBO

is just going to distribute budget. Under here, you're going to have all your ads. A common question I get is, well, okay, we go and put four ads or five ads under here. How many ads should we launch? And when we come up with new ads for this concept, do we launch it in here or do we launch a new ad set? Which is both really good questions. In terms of ad volume, this is really open based on budgets, but you generally want a minimum of three ads and then a maximum of probably 50 to

100, but honestly, the limit is kind of the sky based on your ad spend. And so, I wouldn't really be thinking about upper funnel uh upper limits. I would just be thinking about lower limits, which is a minimum of three. When you then want to go and introduce more ads, this is based on whether you're hitting KPI. So what is going to happen when you introduce more ads under this ads set is you are going to reset the learning phase to some degree because there is sequencing that is occurring across all of these different creative

and then when you go and launch more ads in here you disrupt the sequencing and it relearns how to distribute spend. Now that is not good if the adset's performing well. If this adset is crushing it and you're really happy with it and you're scaling it up, don't go and mess with the learning phase. Don't go and inject a bunch of new ads. you might mess the performance up. And I've actually personally done this years ago. And so that's why I'm very big on not touching stuff that's working. If it's not hitting KPI, and this

is way underperforming, go for your life. Do whatever you want. You can turn ads off. It doesn't matter cuz yes, you're going to break some sequencing, but the sequencing isn't working. So, who cares? Go and launch as many ads as you want in here, right? Do whatever you want because it's not a KPI. So, you need to change something. But if things are working, definitely don't go and launch more ads in here. If you go and make more ads for this concept, let's say the concept's crashing it, so you go and make 20 more ads,

go and launch another ad set, right? And call it concept one and then like shoot two or obviously whatever naming convention you want to use to be able to clarify the difference between these, but then also be able to filter that they are the same concept. Now the question becomes, well, how do you scale? Let's say concept one is performing well. What do we do? Well, number one, increase budgets. Never turn this off. A big mistake that people make is they take something that's winning and they turn it off and they go and launch it

in like a scaling campaign. It's such a rookie mistake just fundamentally speaking because if something works, don't touch it. Leave it. Increase the budget. Now, in Addition to that, sure, you can take these post IDs and you can wrap them around into a scaling campaign. That's fine. And that will work. It works in a lot of accounts. We run scaling campaigns in a lot of accounts. A lot of accounts, we don't run scaling campaigns. Once again, it's dependent on the actual ad account in the business. You can run a scaling campaign. You can take these,

you can post them over here. If it doesn't work, don't do it. Just keep it in the testing. And then next to these two, the only third campaign that I would usually recommend is to have a retargeting campaign that is generally on existing customers. So it gives you the ability to exclude existing customers from here and exclude existing customers from here. So these can purely be cold and that allows you to control budgets. Depending on how long the customer journey is as well, you might want an adset here on 90-day website visitors as a midfunnel.

But this is only if you think that you need to increase frequency on engaged audiences. Now, what are the metrics that matter for being able to measure the performance of creative? Let's list them out in order of reliability. Number one is amount spent. Now, this is counterintuitive. A lot of people look at this and go, why would budget distribution to an ad be the most important proxy for performance over extended time periods? And the reason being is due to the sequencing that we talked about before, right? So, if you have a bunch of different ads,

attribution is only going to go to the last ad. So, if we use rorowaz as the primary KPI, we're going to miss the fact that meta is actually putting all your budget here. Why? When this only has a 1.7 and this has a 2.2. Why would all the budget go here? Probably because this ad is causing purchases elsewhere In the account. And so, this is actually a great topfunnel asset that likely has a low frequency that's cutting through on new audiences. And so, if you're going to look at these three ads, I would say that

this is the highest performer if it has double the spend of the other ones. Even though this is a higher row, sure, but if this could scale, it would have more spend in it. and it doesn't. And so amount spent becomes the best proxy for performance when looking at creative. Number two, sure, is rorowaz or CPA. You just want to be making sure you're looking at 7-day click or incremental attribution. Number three is going to be CPC and CTRs. This is a leading indicator for the effectiveness of the ad. Obviously, you can have good CTRs

and good CPCs, but the conversion rate of the ad can suck and it's not a good ad. And so this obviously isn't a primary indicator of the quality of a creative. That is where you have to use multiple different metrics in conjunction to be able to assess quality. Um but it is a good leading indicator particularly on the fringes. So like if we look at CPCs as an example and let's say this is what your CPCs on average look like across all of your ads. If above this point is, let's say, a $3 CPC, and

none of these ads have ever been profitable, well, we can use this as a leading indicator to know that these ads at very early spend will actually never work cuz it's too expensive to drive traffic. The next two are hook rates and hold rates. These only apply to video assets, obviously, because the hook rate is how many people make it past 3 seconds in the video. And the hold rate, there's different definitions of hold rate. We actually look at a lot of different definitions in our reporting, but as a generalization, this is how many people

make it past 15 seconds. Obviously, hook rates and hold rates are not correlated To conversion performance, but they are a diagnostic tool that allows us to iterate on creative based on what we think is letting us down. And so, we may have a creative that we thought was going to do really well, but the hook rate is terrible against our average. And so, we try to switch the hook out to try to fix that to see if the body actually performs. Same thing applies to hold rate. People might not be holding through the bridge. Turns

out the bridge was actually done poorly. So we redo that part of the video and then we can iterate accordingly. So before I said I would explain verse CBO and the risk profile here. So the reason why you would choose one over the other is actually based on risk tolerance. So what happens when you use CBO is that it's going to play in to Prito's principle. This applies literally everywhere I've ever seen it which is that 80% of something will drive will be driven from 20%. So this applies into companies where 20% of people will

drive 80% of revenue generation. This applies into the ad account which is that 20% of ads will drive 80% of revenue. 20% of ads will hold 80% of revenue. Crazy thing about Parto's principle is that it compounds in on itself. And you see this within every single data set I have done on ad accounts breaking down their ad profile which is that of this 80% 80% of this which comes out to 64% of total is driven by 20% of this which this comes out to 4%. So 4% of ads will hold about 64% of total

spend and total revenue generation within the business. Now CBOS play into Pareto's principle. And what that means is that if you just allow the Campaign to distribute budgets however it wants, it will naturally end up distributing budgets like this where majority of all of your budget in the account will go to 4% of your ads. Now that's fine cuz that's going to yield the best efficiency and performance. But it is super risky because if these ads fatigue, you are in a very bad position. You don't have backups. If we go back to the start of

the video, which is portfolio management, CBOS are terrible for portfolio management, okay? Because you're going to overlever into a few ads. This also becomes an issue when you want to control sell through rates across multiple different products. And so you might have a large product portfolio that all sit under the one CBO and then all of the spend just goes to a few particular products because Parto's principle applies into a product level as well. 20% of products will drive 80% of revenue. And so your entire business will end up skewing and increasing the risk profile

which you might be fine with but that's the trade-off that you have to make. So when you think about these two CBO is increased risk but likely increased ROI. This is decreased risk but likely slightly lower ROI. And it depends on which position you want to be in. I personally, if I'm running the business, I actually choose this every day of the week dependent on like what season and where you are in the business. And so there are periods in time in which I would actually switch over to CBO if I need a quick spike

in efficiency. Efficiency is isn't where we need it to be. I don't care about testing and de-risking right now. We just need ROI to come through. But then ideally you want to be in a steady state over here where the business is derised across more ads across more products and a lot more stuff is working so that if something stops working you aren't in a Terrible position. Uh this next point I'm particularly calling out because we work with a lot of nine figure large retailers here in Australia and if any of them are watching then

they'll know that this problem exists which is that in a lot of large retailers who a lot of my audits are spent on there is this 9010 problem with desire versus painpointled content and this fundamentally comes from old style marketing which is campaign shoots. And so a lot of these large uh retailers in fashion or whatever industry it is, they were built around campaign shoots. And campaign shoots do one thing and that is that they build desire. And so they have models, they have shots that make you want to desire having the product, but they

don't play to pain points at all. And so you end up with the ad account being significantly overweighted towards desire rather than towards pain. And at the end of the day, the best performance marketing assets are those that sit on pain. And so I would strongly recommend looking at your current creative mix through this lens and thinking about how do we actually start to shift this ratio over to 60/40. Now I personally think that the best brands are probably closer to 7030 here. And you can almost think of this in a way of like this

is performance marketing driven assets. This is branding driven assets. Now one last thing in the testing framework is assets for retargeting. So how should you be thinking about retargeting ads on bottom of funnel? This is something I've been saying for a very long time. This is probably one of my first YouTube videos that we put out, which is that on your bottom of funnel retargeting campaigns, you want to think about objection handling. This is the primary purpose of a product ad. And you can think about this mathematically, which Is that let's say you have a

5% conversion rate on the website. That means that 95% of people did not buy. Now the question is why didn't these people buy? What were their objections? Maybe 20% of them had a price objection. It was just priced too high. They never will buy at that particular price. Now, they might buy in the future when you're on sale, but right now they won't buy. 20% of people don't trust the brand. There wasn't enough trust, credibility markers on the website or the ads that they particularly saw to reinforce them to actually purchase the product. 20% don't

like the quality. They think the quality is probably going to be bad. Now, this is arguably a trust issue, but these are really two different things, and they should be come at in two different ways. And then maybe another 20% of people are just not ready to buy yet. Okay? So, it's too soon. Was the first ad they've seen, it was the first ad they've clicked on, they just need a little bit more time to purchase. So, we want to understand this mix. We want to understand why people aren't buying. And then we want to

handle all of these objections in our bottom of funnel created. And so, on the price objection, we want to be showing discounts. We want to be value anchoring. We want to even going for a creative that's like a cost per use angle. These actually crush on bottom of funnel. So if you're selling like a supplement, you compare it to a coffee. It's like half the price of a coffee with the same caffeine, etc., etc. If they don't trust the brand, we want to come in with social proofing. How many customers do we have? How many

fivestar reviews do we have? Do we have any user generated content testimonials that we could use here? If they're doubting the quality of the product, we Want to show UGC of the product actually in use. We want before and afters. We want demonstrations and if it's simply too soon for them to buy yet, we want to be staying present and top of mind with educational or lifestyle content. And then this is what the mix should be in all of our bottom of funnel assets. And this is how you ultimately yield strong conversion rates once people

have been moved through the funnel. The question is, does diversity or volume matter more when it comes to creative? So you want to be thinking through volume through this hierarchy, which is that volume matters the most, then concept quality, then hook strength, then editing quality, then UGC talent, then storytelling all the way down the bottom. Now, it's not to say that any of these components aren't important, but it's the fact that volume ends up creating quality and hit rates are limited within an ad account. And so a way to conceptualize through this idea is that

if you take the best creative strategist, you can pull 10 of them out of 10 different businesses and you can show them five ads that are all objectively good creative and you tell them which of these five was the winner. Which one produced the most spend or the most return? None of them will be able to guess it. It will just be a complete guessing game. They'll assign reasons as to why they believe this creative is better than the other ones. might have been due to the concept quality, might have been due to the hook.

But the reality is is that people are not very good at guessing the top end level of creative performance. And so instead, you want to understand where the bar of quality is and just ensure that ads are reaching that standard. And so if all of these five ads sit above this bar of quality, and we could objectively say that this is a seven out of 10. So if an ad is better than a seven out of 10, it Meets that bar and it is indistinguishable from any other creative in being able to try to guess

its performance. Where you end up dying in creative strategy is trying to win the game of understanding which ads will perform well at a high level. No one is able to objectively do this well. I've never seen anyone be able to pick winners. And so you instead want to become very good at understanding what is the minimum bar of quality. What does a good ad look like? And then how do we reach that level of quality and then maximize volume above it where people get volume completely wrong and the example that I gave way back

at the start which was people were misunderstanding activity for creative strategy is that they were putting tons of ads into the account but they didn't have an understanding internally of what the bar of 7 out of 10 actually looks like. And so when you go and do an ad breakdown of all the creatives that they were putting into the account which was thousands a month the ads just weren't very good. They didn't have any of the elements of a good concept. They didn't have any of the elements of a good hook. They didn't have any

of the elements of formats that were relevant to the stage of awareness of the creative that they were trying to produce. And so because of that, yes, they put tons of volume in the account. Yes, they followed this hierarchy, the most important thing on it. But it was just junk volume. It was just a bunch of ads that aren't even good. And so instead, you need to focus your efforts on how do we understand where the bar is, then how do we maximize volume, and then how do we start steping down through all of the

different priorities to create a good ad. The question off the back of this then becomes, okay, well, how much volume do we actually need then? And there's a really easy, simple answer which I'll give to you. And then there's a more complicated answer that you'd have to go and build a financial model Out for. And if you're an e-commerce brand, uh, we'll give it to you completely for free. We'll put the link in the YouTube description below. So, the simple answer is for every $1,000 in monthly spend, you want one new ad. And so, if

you're spending 30K a month, 30 new ads, 50K a month, 50 new ads. Now this is in AUD also translate this into USD roughly that ad volume is fine. Now the better way to do this is to back propagate based on hit rates and expected spend per ad. So what you can do is you can calculate the average spend of an ad. So if you know that across the last year of ads you've launched let's say 500 ads and you've spent $500,000. Well then each ad here on average spent $1,000. Now, if you know the

average return of an ad, which you could look at either your acquisition me, which is new customer revenue divided by ad spend in the business, or you could just look at rorowaz, but if you're going to look at this, look at it on a 7-day click basis. And let's say your average rorowaz is a three and you've got $1,000 in mean spend. Well, then we know that the average return or you could call this the expected value if we're using probabilistic math is then $3,000 in revenue. So every time we launch an ad, we should

expect on average on a larger data set that we generate $3,000 in revenue. So then we can start to back propagate this into our actual revenue targets. Now more specifically, I would actually delineate this down into new customer revenue to be specific to acquisition. So, if you have a 300K monthly new customer revenue target, you Take 300K, you divide by your 3,000. 100 creatives per month are required. Now, there's a few nice things about this. Number one, you can flex the actual volume of creative in conjunction with spend, which is what you should be doing.

As you go through the year and your forecast changes in terms of expected new customer revenue, your creative volume should flex accordingly. So this should be exactly what your revenue looks like, but also what your creative volume looks like. Now, there are two disadvantages to this simplified model, which is that number one, it doesn't take into consideration the average time of an ad spending. And so these ads might generate $3,000 in revenue, but it might take 4 months. And so that needs to be factored in because obviously volume needs to be a lot higher then.

And it also doesn't factor in the difference between a winner and a loser. If you got 100 losing ads this month, you will generate barely any revenue. If you got a 100 winners, you'll generate like multiple millions in new customer revenue. And so, you need to forecast losing ads versus winning ads separately. And then you need to add the time component. We've built a model that does all of this. Once again, if you're an ecom brand and you want it, you can click the link in the description, uh, fill out the form, we'll send it

out to you. Now, when we're thinking through creative budget, how much budget should go towards production versus actual distribution of the assets, if you're at 4 to 30k a month in spend level, there is no reason as to why your content shouldn't be super founderled. And the actual budget here on production is your time rather than money because you really don't need to go and pay for production on this kind of spend level. This can all be done literally through statics in most Businesses, but even just founderled videos can get you to 30K a month

in spend easily. And honestly, I would challenge that the founder of a business at this size should be creating content because content and creative is the largest lever that exists within the business. And so understanding what good looks like, what good creative looks like, getting an understanding for how to piece it together, what the editing process looks like, what the ideation process looks like gives you the ability to then KPI and train people moving forward on one of the most critical roles in the business. As you then move into 30 to 100k per month in

ad spend, you want to be thinking of allocating 25% of media budget towards production. So at 100K a month, that's 25K a month into creative production. This is way more than like 99% of brands are actually spending. And that's also one of the reasons why brands don't actually scale and do that well. It's because they significantly underweight the allocation that needs to go into their media budget. Now, this is an uncomfortable range because the cost of production is so high excessive to the scale of the business right now that it feels off. It feels like

this is too high. But it does democratize as you achieve more scale. Once you crack through 100K plus, this percentage really should asmmotope down to around about 10% of media budget. And so as a function of revenue in the business, let's say that you're at a 25% ME, 10% of this would therefore be 2.5% of total revenue in the business. So 2.5% of total revenue should be going towards media budget. This kind of spend allocation allows diversity in the creative that you're putting in. This is Enough budget to have UGC going into the account, to

have partnership ads going into the account, to have a full-time either international designer or someone part-time uh in house on statics, and it can likely support a lowcost uh agency in some capacity to supplement with another creative type in here as well. Maybe it's VSSLs or something. I'll give you a real example of a client that unfortunately we had to part ways with about uh 6 to 8 months ago because we couldn't get this concept across to them. And this is why I stress this so much in a lot of content that we put out

now. And it's really important that people understand this particularly through the initial discussions when they start talking with us because we really don't want to onboard brands that aren't across this concept. So this brand was spending 350k per month on Meta when they started working with us. Their CAC was declining pretty steadily over the course of the last 2 years. Now, they only had about 15 ads live within the account on a 350K a month spend, which is crazy. And so, right out of the gates, we made a bunch of media buying changes. The account

was super oversegmented. The previous media buyer was doing so much over-the-top stuff because he didn't have anything to work with from an ad and creative perspective. And so, when you don't have anything that you can do on creative, what do you do? Well, you go to the next thing, and that's let's create 25 different campaigns. let's use big caps, cost caps, let's try everything to try to squeeze these 15 ads, which ultimately isn't the lever that's going to double this business. And so when we did the audit, we identified this as the bottleneck. And what

we did immediately is as we came in, we went and took all old performers and started relaunching and reworking them a little bit. Even though at the time we weren't doing Creative production in house, we just threw this on top cuz we knew that this was the limiter that we had to get this done. So we took the ad account up to about 60 ads. Nowhere near where it needs to be, but this is all we could do. in terms of high quality assets that have been turned off in the past. Now, what the results

ended up being for this brand is we actually turned them around, which was one of the most insane case studies that we've done cuz we did it with no new creative and we actually took them from 350K a month in spend, we increased their spend by about another 40K and we decreased their CAC by about 20 to 30%. And so, new customer revenue expansion off the top of my head, and there's a case study out there on this somewhere, I think it was about like 28%. And so we were able to turn this brand around

substantially because of this. Now the issue was that this founder was very much so of the opinion that this spend on meta is a revenue driver. Any spend towards creative production is a cost center. And so because of that they refused to do any more creative production. We introed to multiple agencies. They went and chatted to them. Uh they actually signed on and then pulled out a of a lot of those deals. And so we weren't able to get any more ads, any more creative. And so because of that, there's only so far that you

can push media buying over time. And so performance started to come down. And then as performance started to trickle down due to these 60 ads fatiguing, ads started to need to be turned off. Spend started getting distributed away from them. And the account slowly fell back to what it was previously. Now, the proposition that we put forward to this brand well before any of that happened was that if you took just $10,000 of this budget, just 10K out of it, and you decreased from Meta and you moved it over to creative production, this could get

us with an agency that we were talking to about 40 assets. Not all unique, some were just hook swap, some were statics, but it would give us 40 assets. This would allow us to almost double the amount of active ads in the account, which would likely unlock at least a 5% improvement in efficiency. Now, a 5% improvement in efficiency on 350K is the equivalent of times 2.5 a 17.5K saving. So, we could either drop 17.5K out of this budget, which we already did. we drop 10 so we've effectively made 7 1/2 grand or we could

keep the budget where it is and this would provide us with an incremental 3x was their return at the time on this so we would make an extra 40 to 50k a month and so you would actually get revenue expansion in this business by keeping spend the same but redistributing it over to production. It's a similar concept to just changing between channels but instead of it being a channel it's creative with an output. But unfortunately, this founder was very set on the fact that they didn't need new creative that we could just repurpose these existing

assets and that would continue the business going. And this all definitely could have been fixed well ahead of time simply through a slight allocation of this budget over into production. So definitely avoid being that brand that sees creative as a cost center rather than a revenue driver. Now, one last tactical piece of advice here that applies right now as of April 2026. This might change as the algorithms change. This might not apply into the future. But one of the best ways to quickly test high volumes of assets right now, we actually do this on all

of our own ads, is that you post Them as trial reels and you cut the CTA off. So if there's a call to action at the end, cut the call to action out, throw it up as a trial reel on the business page that's going to run through or the personal page that's going to run through. And then what you will get back is two things. Number one, you'll see how viral does the post actually go? Does it get any views? Does it not? So you're going to have comparative data. But number two, you're also

going to get retention graphs, which is going to show you how you've retained viewers throughout the ad. Now, the retention graph is going to show you, was the hook bad, was the bridge bad, was the body bad. And this lets you iterate quickly on the next shoot without needing to actually push the ad straight into the ad manager. And you can do this before it's perfected. So, if there's some issues with the captions or there's some stuff that isn't great and it needs to go in for another round, you can just take the V1, throw

it as a trial reel, cut the call to action off in app and get feedback while that round two version is actually getting worked on. And then you could pass even more feedback on during that round two within 24 hours before you've even got the final asset. And so, we found this is a really effective way to take bulk amounts of creative and quickly test particularly on hook variations as well. So we can take like 20 hook variations, throw it up as trial reels, what hooks actually perform the best, and then only put five ads

into the account rather than wasting 20 ads and spending money on trying to figure it out. So now we move into creative fatigue. So there's two concepts to understand with creative fatigue. Number one is the two reasons why creatives fatigue. And the other concept is to understand the maximum spend threshold of an ad over time. So when you think about launching an ad and on the y- axis we'll put spend and on the x- axis we have time. You launch the ad, it does well. And so, one of two Things happen. Either you start manually

spending more on it, so you start increasing budgets, or Meta will go and automatically distribute spend to it. And the spend goes up and up and up and up. Eventually, there will be a limit in how much daily spend that ad can hold. You'll reach that limit. You'll probably go past it. You'll correct. And then you'll find the maximum daily spend limit of the ad. Let's say it's at like $500 a day or $1,000 a day. From there, this ad will continue spending moving forward in time. It will eventually start fatiguing. It will start falling

off. You'll either start pulling budgets or the CBO will pull budgets out of it and then eventually you will just turn it off entirely or Meta will stop distributing spend. Now, this area under the curve here is the total spend capacity of the ad. So, you can go and do derivative math and just get an understanding of how much area there is there and you will have total spend capacity. Now, what's important to understand is number one, every ad has a total spend capacity. Every ad will fatigue. Every ad will get turned off. Number two

is you can change what this graph looks like over time. And so, if you want, you can squeeze ads harder in the short term, but they will fatigue faster. And so, you could take this ad all the way up to here in daily spend, but it just means it's probably going to fatigue and be turned off here. Or you could go even more aggressive and you could actually launch it at a high budget, take it way up, but then it's going to fall off. Or you could obviously do the opposite and you could under portfolio

management theory, which is what we did at the start, is that you actually just want to introduce this ad But keep it at $100 a day and you'd rather have it live for 6 months than die off within 6 weeks. Now, this is ultimately more so going to be a product of the inventory constraint and demand within the business. So, if you have a lot of inventory that you need to sell through for the current profitability position in that month, then obviously you should go with option one here and just squeeze the winning ad to

sell through as much product as possible, even if it's at a degraded me. On the other hand, if you don't have much inventory, um, or if you're waiting for more inventory to happen or if you're growing too quickly that your cash conversion cycle can't actually uh, keep up with the growth rate, well, then why would you do this? This would be a ridiculous decision. you would instead want to operate at higher efficiency and let the ad breathe for 6 months. However, if you're working with an agency, they don't have any of that wider context within

the business unless they're asking for it, unless you're actively giving it to them. And so, they will see an ad that's winning and they will go, let's squeeze the hell out of it because of two things. Number one, we just started working with this brand and this will give us a great 30 to 90day case study. And number two is, well, this is giving us the best rorowaz and we report on rorowaz or efficiency and profit and so let's do this. But the issue is it completely screws up the long-term jevity of the ad account

and the business because you've over prioritized short-term gain for the long-term stability in the ad account and the inventory purchasing. So you're just going to go out of stock here and you're going to ruin a lot of what's going on. Then the two drivers of creative fatigue. Number one is the ad itself just become stale. So this is a function of it's just reached its maximum spend capacity and it will Ultimately just achieve at the end of its life cycle. Um, this could also be just a function of the ad has the same visuals, it

has the same concepts, it's targeting the same audience as the other creatives in the account. And so you'll get much more faster fatigue because you're targeting the same audience with the same value propositions. The second reason ads fatigue is that the audience is actually just too small. When we went through the persona component of uh this video, we talked about how specificity in the persona is critical to being able to script a better ad that resonates with a very specific target demographic. This is also going to generate way better results because you're talking to a

few people rather than a lot. Making very good top offunnel unaware ads is incredibly difficult. The skill ceiling is very high. Most people shouldn't try to do it unless you're already a $50 million plus brand. Now with that, if you speak to a very small persona, there is obviously going to be a maximum daily spend capacity that you can spend talking to that persona. And so if we go back to the example before of men trying to run sub 4-hour marathons that have gone to the gym, that are trying to cut 5 minutes off their

time using creatine. Very specific target demographic. Now, there's still a lot of people there. A lot of people are trying to run marathons and trying to run sub4, but you're not going to be able to spend, particularly in the Australian market, $10,000 a day on that creative. It's just not going to happen. And so, you need to find out what the spend capacity is, keep it there, rotating creative so that once these fatigue, we can continue to hit that audience over and over again because that is an audience that churns, which is really important. So,

people will enter into that audience, then run a marathon and enter out of it. So there's constantly new people coming in that we can continue to saturate, but we don't want to oversaturate, over Prioritize, overspend, drive up frequency, and then just ruin the profitability of us targeting this audience altogether. Now, an obvious key learning here is that broader audiences will scale much further. And so if you have a top offunnel unaware ad that speaks to a much larger TAM total addressable market, you will obviously be able to spend a lot more money on that. In

a B2B context, an example of this is that if I have a persona call out at the front of one of our ads that says e-commerce CFOs in Australia owning over $und00 million a year in your business, that is such a tiny pool. There's like how many people fit that particular persona in Australia? It's probably a thousand, maybe 2,000. And so because of that, when I go and put that ad in market, I can't spend a lot on that ad. There would be no point in me going and putting $10,000 a day behind that creative

because I would then just be hitting the thousand people like 600 times a day. Makes no sense. But then if I expand the persona call out to e-commerce founders earning over 100k a year, well then that's a much larger target demographic. So the daily spend capacity improves. Now does that mean I should just do broad callouts? Absolutely not. Because broad callouts means you have to have better creative that resonates with more people. And so you're actually better off having the more specific persona call outs that force you to create better assets and then move into

broader audiences once you hit a particular scale. In terms of tactical things that you can do when creatives fatigue, there are changes that you can make. Number one is you can make iterations. And so you can take an asset that's already done well, which I Mentioned before, a hero ad that started to fatigue, and you can just rotate new hooks on, and you'll be able to get more spend through that asset. Number two is you can move the ads over into a cost cap campaign and you can try to get more spend through here. Now,

this will also work. It will fatigue the creative faster and it's a short-term tactical media buying strategy. It's not something that's really going to materially change the business. Number three, you could go and duplicate adsets. It's not something that I would recommend. I'm not even going to go and put it down here. I know it's something that people still argue works. I think it's too tacky as a medium buying strategy. It used to work back in the day and I could explain the concept as to why it works and why it doesn't now. But if

it's something you want to do, you can do it, but I wouldn't recommend it. You can take the post ID of the creative so it retains all of the engagement and you can go and move that into another campaign like a scaling campaign to try to get more spend through it. This sometimes works, but once again, like this isn't going to save the account. And then you could rotate, which is kind of a variation of number one. You could rotate formats. And so if this is a video that's doing really well, you could try to

translate it into an image. If it's a VSSL that's doing really well, you could try to translate it into UGC. If it's an image that's doing really well, you could try it translated into an unaware long form copy static image, which is a completely different strategy. So they're really the four options. As a creative fatigues, it is just worth noting that once a creative fatigue fatigues, there's not like a whole lot you can do about it. you know, that's why you need the portfolio management approach and you need other ads that are doing well so

that you can fall back on them. There Are things that you can do though just to squeeze a little bit more out of the back end of a creative, which is what these tactics are. Tying all of that back into portfolio management on accounts spending greater than 50k per month really the job of the media buyer is portfolio management across the ads and diversifying risk here portfolio management against the products and assuring commercial objectives within the business is aligned with what is actually occurring within the ad account and so the 8020 parto principles rule applies

into ads and so the question becomes do we want this naturally occurring within the account or do we want to force back against it through introducing some structure so that we're not overleveraged on ads. Same thing applies into products and then on commercials this becomes are we making a balance sheet or a P&L player. And what I mean by that is that agencies will always optimize to maximize profit. Okay, the idea is how do we have the highest efficiency? How do we have the highest profit contribution? That is what they are KPI on. But often

in a lot of businesses, if you do that, if you overoptimize towards profit, you end up end up underoptimizing towards the balance sheet and you end up in a position where you have a lot of unsold inventory sitting on the balance sheet that isn't selling through. And yes, you are making profit and yes, you are spitting out cash, but the cash gets tied up in the balance sheet quite quickly. And so you could have a P&L that says $2 million in profit, but then you have a balance sheet that has $3 million in unsold inventory,

and this business is actually negative $1 million in cash because all of this cash that they produced just moved into unsold Inventory. And this is fixable through paid media through the account structure. You just need to restructure the account to sell through this stock. It's not going to give you the best return. Your efficiency isn't going to be that good. You're going to have to make creatives around this stock as well, which is going to be a hard pill to swallow, but it's going to allow you to get out of this position and rebalance the

business. And this is ultimately what the media buy needs to be across. If the media buyer isn't across inventory, that is not good. Creative strategy is worthless without the production. And so, moving into creative production, teams, briefing, and the process. I want to start with the three content types. Number one is winning content replication. So this is taking your winners and replicating them in some capacity. Okay? Whether that's iterations on the hook, whether this is translating them into different formats, you're taking concepts, you're taking ideas, you're taking creatives that have already worked in the past

and you're replicating on them. This is by far the highest impact content types that you can do across paid as well as organic. If you have organic content that's performing well, what most people don't do enough is just repost the same stuff, redo the same stuff over and over and over again because every single time you do it, it performs just the same, but it reaches a new audience. So, this is where a lot of your effort should go. Then you move into iterative content. This is trying to fix underperformers through changes in hooks, changes

in formats. You're working on tangential creative ideas. So, if you have a concept that's working, you're making a slight change to a new persona, or you're introducing a new offer and you're trying to get it to work, this is the second highest priority, prioritized by leverage. And then down the bottom here is net new concepts. This is testing completely new formats, completely new angles, completely new Concepts. You should be doing this and you should be allocating time and energy to it. How much you allocate is based on the performance up here as well as your

risk profile. What I would not recommend is some people do 0% here. This is a terrible idea because all of this will eventually fatigue and fall off and you'll be in a very terrible position. So you need to constantly be finding new stuff to then flow up to the top of the funnel here. But I would say that most people actually overweight towards net new concepts. I would say that I actually do this myself in my own organic content, which is that probably 80 to 90% of the content I put out is net new. It's

new concepts. It's new ideas. It's new stuff I've never spoken about before. And 10 to 20% is just the exact same stuff over and over again. And what ends up happening is all of the winning content replication always does unbelievably well, holds 80% of the views. And this stuff doesn't really work that well. It does allow for continuous winners to be pushed up here, but this is probably an overallocation. And so my recommendation generally speaking, and this is going to change contextual to the business, is that you want about 20 to 30% of effort down

here. another 20 to 30% of effort down here and about 50 to 60% effort up here. Now, when it comes to briefing in creators, the key here is that the quality of the output of the creator is going to be a product of the quality of the brief. Okay? So, you want to do two things when you're sourcing creators. Number one is you want to check what they've made in the past. You want examples, and if you're not doing this, that's kind of crazy that this is kind of self-evident, but you want examples of what

they've produced in the past. But not only that, this is kind of where the additional source is, is that you go under that and you get the brief of the example that they created because that's going to give you a direct one for one comparison of how does this creator produce against a brief. Was the brief super in-depth and therefore we need to match that level? Did this creator make something amazing with a terrible brief? Okay, then we know that we can give them relatively broad strokes and they're going to pull it into something that's

going to be good. Now, generally speaking, I want the highest quality, most inep in-depth brief possible. If they don't want to hold to the script, that's fine. Shoot one that's held to the script. Shoot one that feels more natural. We can choose the best. We can run them both. Um, but you really want examples and the briefs associated with those examples to be able to set yourself up for success when you're going and building out creative briefs. When it comes to building out teams around production, I've seen a lot of different teams that all produce

enormous amounts of volume, but there's a few different key hires that you should think through when you're thinking about maximizing volume within a team. Uh, number one is a video editor specifically for video ads. Now, if this person can be AI native or someone that uh has built a lot of AI creatives in the past, this is even better because this just unlocks the ability to make way more assets and work with way more footage that doesn't actually exist yet. Another one is a designer for statics. Okay, now both of these can be based anywhere

globally. Both of them can be part-time contractors based on actual outcome or they can be full-time within the business. Then it has a third. You want a creative strategist. Now, this is a new role that didn't really exist two To three years ago and therefore there aren't actually many creative strategists out there. So, what do you actually look for when hiring for this role? Well, you you look for number one, copywriters, because creative strategy is primarily copywriting. They are writing the scripts. They are writing the brief for the static image. They are ideulating on new

personas. This is exactly what copywriters have been doing, except copywriters are now redundant because LLMs will just produce most of it. And so these people are coming up with the ideas and then they're moving this into a claude skill which needs to be built by you and brand purposed to then output an actual script that then goes into production. The next option is you just poach a creative strategist from somewhere else. There's enough floating around now that you can probably find one. Or the third is you actually backfill this role from a designer or a

video editor. And so you train a designer or a video editor up into becoming a creative strategist. And this actually allows for faster feedback loops because they can ideulate on all the ideas and then they can design and then they can obviously brief in for any video edits. These are really the two pathways into creative strategist as it currently stands. Now, in terms of a side note here, what a lot of people will do with their creative strategist, which depending on the scale and volume that you're at, depends on whether you want to do this

or not, is they will actually pay out a percentage of ad spend based on the creatives that they obviously come up with and then brief in. And so the creative strategist is directly incentivized to maximize hit rates because if their ads don't get spent and don't perform, they don't get paid. Typically, this sits at about 3% of ad spend. So you'll pay out the creative strategist on top of a base Salary with a percentage variable model that is obviously directly proportional to the results within the account. Um, now I've seen this work really well on

mid-size brands. When you get really big, this gets outrageous because these creative strategists end up getting paid like $80,000 a month. And so then the model kind of breaks and you end up moving away from it. A really important part of this whole component of all production is where does AI fit into the mix. When you break down the creative process end to end, you've got ideiation as step number one. You then have briefing. You then move into the actual production and then you move into the publishing into the account and ultimately the analysis process

that feeds back into here on future ideation. Now, a lot of people are going and applying AI and they're building Claude skills for this right here, the ideation process. This right now, at least at the time of recording, is not the place where you want to automate because the reality is is that all of the LLMs just output the meme. And so no matter how much context you provide, no matter how much documentation you have, no matter how much you build out your skills to go and scrape Reddit and all of these different forums to

come up with unique ideas, ultimately the ideation process here ends up not being strong enough to differentiate from competitors. And you still need a creative strategist to really do a lot of heavy lifting here to validate the concepting. When you then move into briefing and script writing, this is when this can be heavily leveraged with AI. That's when you want to build a claude skill to be able to walk you through like a 20 questionnaire about who's your persona, who's the concept, who's the offer, do you want this to be long form, what stage of

awareness is it at? And so you're answering all of these questions and then ultimately at the end of it, it Pumps you out a script or a brief that's going to be 99% of the way there and you just have a 1% tweak. On production, there's a lot of leverage right now in AI as well. It's obviously going to speed up workflows, but in VSSLs and static image ads as well, you can pretty much get away with AI for 90% of it. Then in the analysis process, this is also where AI becomes a little bit

dangerous because it won't overweight to the hierarchy of metrics that you're looking at and won't sit towards the importance of the account structure and aligning the context of the creative to the commercial objectives of the business. So there is leverage here. You can use it in analysis. We use it in analysis. uh but there was so much context required that it still misses a lot of the time and if you're using AI analysis to make core decisions within the business a lot of the time it actually misleads you in the wrong direction and so with

AI application I would try to skip the outsides and really maximize throughput on the middle want to make a quick note here on content first brands which is brands that are starting with organic content first before they even touch paid I think this is the approach for 26 and 27 because it's not due to the fact that organic content outperforms paid, but it's due to the fact that organic content forces you to become good at creative, which then compounds into paid when you know how to make content that people actually want to watch. And so

my recommendation to smaller founders that are watching this video is that I would be posting organically every single day. I would be getting good at figuring out what good content looks like. See what actually gains views. Look at the retention graphs. look at the share and save rates because this is ultimately Going to improve your ability to make ads that are actually at a good quality. Another side note here is that when it comes to creators, you want to make sure that they're regional to where you're actually targeting. So, as an example of this, in

Australia, you want Australian accents. When you go to the US and you try to translate an Australian accent into the market, it just does not perform as well. So, you need native US accents. uh UK to AU actually translates decently well, but still native accents always outperform. So, you want to make sure the creatives are native to the actual region that you're targeting. It can be the exact same script. It just has to be a local creator. And then one final recommendation when it comes to production is on copywriting. Copywriting is still one of the

highest leverage skills when it comes to building out video asset scripts and also building out static images, particularly long form copy static images. There are four books on copywriting that I have read and I recommend you read all of them if you're wanting to become good at creative strategy and copywriting. Number one is Breakthrough Advertising by Eugene Schwarz. Number two is Scientific Advertising by Claude Hopkins. Number three is Oglevie on Advertising by David Oglevie. And then number four is influence the psychology of persuasion. If you read those four books, you will become exponentially better at

being able to actually script out copy that will convert better through ads. So if we now go and put it all together, the foundation is that you want to understand how the ad platforms work. Understand that the creative is the targeting. Concept separation is what's going to allow you to reach new audiences. Then when it comes to concepts, you want to understand how to split up a concept. Start with personas, move into angles, then move into the offer. And as the final component, you want to bring in the format. Hooks drive 80% of the view

through within an ad. So it's really important to spend time in optimizing that portion of the ad and usually hook testing across multiple different videos. The three components of a hook is the visual, the copy, and the audio. You can change any three of these variables and improve performance. Then you move into the format. You want to start with the concepting first, then always move into formats. Statics are great for quick testing on new concepts, and then they can be validated and moved into videos. Statics also are incredibly underrated, and most people don't think they

perform well because it's a skill issue. I have seen static image ads spend $500,000 in lifetime spend. And you want to always be careful on DPAs. DPA are good, but they generally overspend. and the rorowaz number over inflates them. When it comes to testing, you always want to be testing one variable at a time. You want amount spent to be the best proxy for performance, then rorowaz and the tertiary metrics. And the actual structure of the ad account should be a product of the complexity of the business. When it comes to thinking through volume verse

quality, you want as much volume as possible as long as it meets the bar of quality that we would classify it as a good ad. If it's a good creative, no one will be able to guess which one will win. So, put them all into the account. When ads fatigue, it's because you're either trying to push too much spend through a niche audience and it's driving up frequency, or the ad has simply reached this lifetime spend, it's Falling off, and you need to diversify by thinking through portfolio management risk. When it comes to production, we

ran through all of the relevant team structure that you should think through as well as the fact that you want to be putting majority of your production cost towards iteration rather than net new concept design because that's going to drive most of the performance. And then if we do one final breakdown of spend per month and the recommendations on how you should approach creative at 4 to 30k a month, the founders creating the content, it can be done on an iPhone. You want it to come across natural. All the leverage is going to be in

how strong the scripts are. You want to be working on three core concepts. You really don't need more than that at this scale. You just want to do these well. You want a high testing allocation. Realistically, 60 to 80% of your budget's going to go towards testing because you just don't have enough spend yet to even scale winners. And then you can use trial reels as well as a hack for validating ideas. Once you move into 30 to 100k per month, you want two to three UGC creators. 25% of this spend needs to be going

towards production. You're going to move into five to six concepts. You want to start branching out and d-risking. You're going to have an and then likely a scaling campaign. You're sitting at 30 to 100 new creatives per month here or you're using our creative calculator and it'll give you an exact amount. And then as a little hack at the bottom, you want to be introducing partnership ads. And then if you're at over 100K per month in ad spend, you really want a full creative team. Someone who's doing scripting. You want an editor. You want four

to six creators. And you want a designer. Now, these could be part-time, but you want these resources in place so that you have the production machine in place to Churn out creative. That's really what you want to be looking for at this ad spend level is you need a system that consistently produces creative on an ongoing basis. If that isn't in place, you're just constantly going to be scrambling and creative will be the bottleneck. You want at least 10% as a minimum budget going towards creative at this spend level. You want 8 to 10 active

concepts within the account. You want at least 100 new assets per month based on the basic 1k per ad formula. And you want a financial model in place that's tracking your cost per asset, ideally across the different sources. So, how much is these creator pieces costing you? And then what is the average profit contribution per ad on these creators? So we can start to see if the cost per asset of these different content types is actually worth it. And you want to be tracking ROI at an individual ad level based on ad spend against the

cost of production. For most people right now, creative is the constraint on meta. You don't know what good looks like until you start doing volume, building out concepts, scripting, getting ads in the account, and beginning that feedback loop. So, as a wrap-up, if you're a performance marketer and you have at least 2 years experience, please reach out to us in the description below. We're always hiring. And if you're an e-commerce brand or a retail business doing at least $5 to $10 million a year at a minimum, you can also click below or reach out to

us and we'll do a completely free audit where everything that you've learned in this video. We will actually translate directly into the ads that you're running directly into your ad account. We'll pick it apart so that you can know how to unlock the current bottleneck that's stopping you from growing. And also, please subscribe. Most of the Google Ads playbooks running inside agencies right now are 2022 2023 strategies. They haven't been updated for the massive Platform changes that have occurred over the course of the last 2 to 3 years. Almost every part of the platform has

shifted. Google tries to ship a new product every 3 to 4 months and it constantly changes the landscape of how you should be approaching the platform. In this video, we're going to be breaking down how the platform works in 2026, what structure and segmentation should look like in your account. We'll go through PMAX. We'll go through where standard shopping still has a place in 2026. We'll talk about smart bidding and bidding strategies. We'll go into the GMC feed and we'll rank everything that matters in the order of importance. And then finally, we'll give you a

90-day playbook at the end, which is a 90-day plan that you can roll out on your Google account to be able to transition it into a position that's fit for 2026. The quick commercial frame of Google ads that's really important to understand is that almost everyone watching this video is either overspending or underspending. Now the overspend is very common and it comes from the fact that Google sits very bottom of funnel and all of the top offunnel generation in the business is actually coming from platforms like Meta, Tik Tok, affiliates etc. And then Google just

captures all of the bottom intent demand and then claims a really high rorowass. And so as a function of that, a lot of people just push up spend on Google despite it not being that incremental in the business and they end up with overspend. Then there's the other side of accounts where there are people that understand this idea. In fact, they're very indoctrinated into it and they see Google ads just as a bottomfunnel platform and that's all it is. And so as a product of that, they don't take Google spend over $10,000 a Month despite

them spending a4 million a month on Meta because the logic is well yeah Google's bottom up funnel. I don't want to scale it. Everyone overspends here. Why would I spend more? But they're in a category that is very scalable on Google. There is tons of cold traffic every single day searching for relevant key terms that their product could serve under. But because they don't trust the platform, which fair enough, they don't go and spend on it and they don't go and maximize it. So, it's really important to understand that you're going to be in one

of these two camps. And by the end of this video, you'll understand which camp you're in and where you need to start moving spend and structuring spend in the account. There's really been five big shifts over the course of the last few years in Google that you need to be across. Number one is that Pmax hit its ceiling. So, when Pmax first launched, it was a pretty promising campaign. In fact, Google was likely giving discounts at auction so that people would preference into the new product. And so when you used Pmax, you got ridiculous results.

We saw some crazy results on accounts 3 4 years ago. Then everyone started to weight all their spend into Pmax and the arbitrage opportunity there kind of went away. You also have no clear visibility into where spend is going. And then what you end up finding is that as we scaled performance max campaigns on a lot of large accounts to 100 200 $300,000 a month in ad spend is that we didn't start to see the returns show up on backend revenue. And this is where we started to dive into the true incrementality of the platform

to understand, okay, even though Pmax says that we're still at a five rorowaz and We've doubled budget, why hasn't backend revenue moved? And that became a real core testing framework for us to start thinking through redistributing budgets out of PMAX into standard shopping. Standard shopping pretty much was phased out of a lot of accounts uh when PMAX started to first arise, but now standard shopping is back and it's because once you scale Pmax far enough, you just see diminishing returns and you can't scale it any further. And so if you want to continue to grow

cold new customer acquisition on the platform, it will require standard shopping being layered into the structure. Number three is match type definitions have changed. Exact match in search campaigns isn't exact match anymore. It now matches to the same meaning. And so there's actually flexibility. Phrase match got a little bit broader as well. And then broad match is just as broad as it's always been. And so this affects a lot of the old strategies that people used to do. For example, if you've been running Google ads for a while or you might still have this in

your account, SCAGs was a really common strategy back in the day where you would have a single keyword ad group for every keyword and you would monitor at an ad group level. I wouldn't recommend running this anymore. In fact, this stopped working in like 2020. Now, you might have an account where it works and therefore you've just continued to run it, but there's a better way to do things. And so, we'll dive into that later on when we talk to search campaigns. Number four is that Google has introduced AIMAX. They've also introduced demand genen. And

then number five is DSAs are being deprecated. So, on account structure, first rule of account structure, consolidation will always beat segmentation. And this is a product of how the account structure works in segmenting and siloing data. So let's say you have three campaigns. This is Three shopping campaigns. And let's say it's all the same products as well. You're just testing different bidding strategies or you have some kind of strategy going on here. What happens is that this top campaign maybe has 50 conversions every 30 days. This next one has 20 and then this next one

has 10. Now the issue here is that all of these campaigns learn in isolation. Now there is a little bit of platform level data sharing but just ignore it for now to simplify the model down and just imagine they are completely siloed out. So these 50 conversions are teaching this campaign who to optimize for but they're not getting shared with this campaign and it's not getting shared with this campaign. So they're isolated out. What that then means inherently is that this campaign down the bottom here with only 10 conversions isn't going to perform very well

because it's modeling off 10 conversions. It's using the psychographic data points of just 10 users to decide how to adjust bids in real time on future search terms, which is crazy. And so because of this, this campaign inherently won't perform that well. And you'll often end up finding this when you see a cam account structure that looks something like this is that as you go down in the conversion volume, you also go down in return on ads. And that is usually causally related to the volume going through the campaign. And so how we could immediately

get better performance in an account like this is we simply delete campaign B. We delete campaign C and we move all the products and the budget up to here. And now in the top campaign we have 90 conversions per month rolling through. Because this is such a larger data set, it will be able to model to Users better, create more accurate bids at the auction, and therefore produce a higher return on ad spend than if we were to segment out. And so the question becomes, why do we even segment out? Why don't we just always

run like one PMAX campaign and that's it? Because that's how we'll see the best ROI. And the reason being is that unfortunately, most businesses aren't that simple. And the complexity of the business needs to match the complexity of the ad account. And so if you have as an example here three different product categories, one of them is grade C inventory that we either have to burn discount by 80% or give away. Well, do we want that getting grouped into this single campaign and getting no ad spend? No. We probably want a little bit of spend

behind those products so that we can turn them over so we don't have to put them on heavy discount. Therefore, we launch a grade C inventory campaign here that doesn't do very well. It doesn't hold many conversions, but we don't care cuz at least it's just moving over a little bit of stock. We then also might have new arrivals that are constantly coming in. Now, we want that surfaced right at the top of Google Shopping because we want to not only turn over our new arrivals quickly, but we want the seasonality of our ranging to

be relevant to today. We don't want super old products that are no longer relevant to the season to be popping up on Google Shopping. So, as a product of that, we have a new arrivals campaign. And then lastly, we have the core campaign at the top here that maybe we started with when we were a smaller business. And then as we've scaled and grown, this has just grown in ad spend. And this is why the account is structured in this way despite it not meeting the rule of consolidation over segmentation. And so the real key

here is that you always want to be as consolidated as possible whilst introducing commercial realities into the attic. This is fundamentally why an agency needs to intricately understand your business down to the grade C inventory level or else they will just segment with no reason behind it and just decrease performance for the sake of adding complexity into the ad account or they will do the opposite and they will consolidate up under this rule. However, you would actually be able to better align the ad account with the commercial objectives of the business if you did have

a little bit of segmentation. So that is how you need to be thinking through this problem within the account. A good way to think through this problem is that the goal is always one campaign that's never realistic but it's the default and then every additional campaign that you layer into the account needs heavy justification. Now the justifications that count is number one brand versus non-brand. So you don't want your one Pmax campaign to have brand searching. So, if you want placements on brands, but then you also want cold, this needs to be two separate campaigns.

And I would always recommend this. Number two is geography. So, if you're selling in multiple different countries, that should always get segmented out at a campaign level. Number three is different margins or efficiency goals. A 70% collection in terms of gross margin Cannot be bundled with another collection that has 35% gross margin. Because if these share the same bidding strategy, let's say you're just running a 300% target rorowaz. Well, this is like barely profitable at that rorowaz. This is incredibly profitable at that rorowaz. Now, because Google, at least with most people's setups, doesn't have visibility

into the margin profile. It will just optimize towards revenue. And if this is generating more revenue, this is where all your spend goes despite it not even being profitable. And so you need a degree of segmentation in the account based on the margin profile. And then number four is product categories. And the reason being is that you can have very very different products and they don't belong together and they don't belong learning together as well in the account. As a super extreme example, let's say that you sell office chairs and then you also sell plants.

Now yes, you can buy plants for an office. They have some kind of similarity and that's probably why you're selling both products. But the reality is plants are going to learn towards a very different audience than your office chairs are. And so as a product of that you do want the data segmented so they can learn in isolation to their own unique persona. Now if we quickly talk about single keyword ad groups you do not want to be putting one single keyword into an ad group anymore. It does not play to the way that the

accounts want to be structured by Google. Now, Google and Facebook say a lot of stuff because they want you to spend a lot of money and their advice isn't necessarily what you should be doing 80% of the time, but there is method in 20% of the stuff that they say. If they say consolidating keywords Down is beneficial and then you test it and it works well or we can be pretty confident that consolidating keywords down is beneficial. And that is what we see constantly across all the accounts that we work on. We have still tried

SKAGs in fact because we onboard clients that have this running in the account and then we transition over to single topic ad groups instead and we always see better performance. So let me give you an example and I won't give an e-commerce example. I'll give a service-based example as search campaigns are way more uh relevant to service-based businesses. Quick side note here. If you're an e-commerce brand, we only work with e-commerce brands. All the content is about e-commerce brands. video is really orientated towards e-com and retail businesses. In e-com and retail, you really want 80

to 90% of your spend to be in shopping, not in search. And the reason for that is that you always end up with higher quality traffic and higher intent audiences on the website. It is due to the pre-click information. When you look at a search ad, what do you get? You get a headline, you get a description that no one reads, and then you usually get a couple extensions that can help. So, if I go and search for buy red dress, size medium for wedding, I get a search ad and it says red dress for

wedding size medium. Now, it matches my intent. It tells me what it's going to be, but I have no idea what the price is. I have no idea what it looks like. I have no idea what the other options are. I'm very limited. And so, I'm trusting that when I click on that search ad and charge the advertiser $1 to $2, that whatever's going to be on the other side is what I actually desire. And in fashion particularly, it is very visual based. And so that is generally not a good idea. It's not a good

idea to just run ads and say, "Hey, we sell dresses." Because you'll end up with a lot of People coming in the door and a lot of people go, "But this isn't the style I like." The great thing about shopping ads is you show the user an image of what the product looks like, the price, the title, potentially any promos or reviews that are currently active, the brand name, and then any other extensions that drop down. On top of all of that, five of your competitors pop up right next to you. So the user does

a comparison across six listings and chooses the one that's most relevant to them. So they're already pre-bought into your listing being the best. Now on search, there's almost no information. So you get worse conversion rates, which is why shopping will always outperform search in pretty much any e-commerce business up to the point until you completely saturate the shopping network and then you go into search for additional volume. Back to single topic ad groups. What this looks like is we have three ad groups. We have general plumbing, emergency plumbing, and then affordable plumbing. Where this previously

on a single keyword ad group set up, you would just have phrase matcher, exact match for these key terms, and that's it. This whole ad group is just bidding on the key term general plumbing. Instead, this is a topic. And so under this sits 10 to 20 themed keywords. So all keywords that are relevant and themed to this topic. And then the same thing for emergency. And then the same thing for affordable plumbing. And then the ads here are tuned to whatever that theme is. So they're contextually relevant to all the keywords that we're bidding

on. I'll tell you why hyper segmentation within Google doesn't work anymore. And why scaggs has turned into stag, which is I used to run scags back in 2019. This was Actually the first Google campaign I ever launched on the platform was running a SCAG campaign. So I'm very familiar with it. The reason why this used to work and for new performance marketers or new people running brands, you won't even remember this, but ads in Google didn't used to be responsive ads where you put in 15 headlines and four descriptions and tons of images and then

it will just dynamically adjust what headlines serve to the user. Instead, back in the day, you would provide just three headlines, just one or maybe two descriptions, and that was it. And there was no ranking or sorting them around. You would choose which one's the first headline, which is the second, which is the third and then that is the ad that served. So you would choose exactly what the ad looked like. You would choose the exact keyword it would place on and that would be that makes total sense, right? We want our ads to be

contextually relevant to the exact keyword that we're placing on. And that's why SCAGs existed. That's why it did well. But now that's not how it works. Now you create a responsive ad. And a responsive ad has 15 headlines. It has four descriptions. It has a bunch of other asset variations that you put in there. And because of that, there is 2,700 possible combinations of this one ad that can serve to a user. And so Google is going and split testing all of these headlines, all of these descriptions, all of these assets against each other in

different combinations to try to figure out which of these 2,700 combinations is going to perform the best. Now, to properly test this, you realistically need to serve this ad against about a quarter million impression. Now, are you going to get a quarter million impressions on this ad If you're just bidding on one keyword with exact match in a SCAG setup? Absolutely not. And so, the only way to get out of learning and to get the campaign understanding what combination of an ad should be served to a particular audience is through higher volume getting put through

at an ad group and an ad level. The added benefit too is that because this is dynamic, it can just dynamically adjust based on the different keywords. So, we don't need to set up one ad for this keyword, one ad for this keyword. We just set up a large ad and then Google dynamically figures out what headlines and descriptions work based on which keyword it is within the themed group. A quick note on account hygiene, you always want a proper naming convention set up just so someone can easily jump in the account, have context, understand

what's going on. The naming convention that we like to use is obviously the agency tag at the start so we understand what campaigns we launched and what anyone else might have launched. the region, campaign type, is it a PMAX, is it a shopping, is it a search? The audience, is this a cold campaign, is this a brand campaign, is this a theta campaign? And then the goal at the end, which normally we'll actually just drop this off cuz the goal in 99% of the case is just sales. It's conversions. But in some cases where we're

doing some different campaign testing, uh, this might change out and therefore we'll specify. I now just want to rapidfire you a bunch of mistakes that I see in Google accounts all the time. Before we dive into the large section of the video where we start breaking down account structure, performance max campaign, shopping, etc. So, number one is G4 being used as the primary conversion. You shouldn't use G4 events in the account because there's About a 10 to 15% sometimes even higher attribution gap. So, it misses about 10 to 15% of the conversions that would get

tracked if you just had direct snippet code installed on the website. Number two is just running one asset group on a performance max campaign. We'll go into why that's a mistake later on, but always make sure that you have a degree of segmentation within the PMAX under the asset group level. Next one is running broad match with the maximize clicks bidding strategy. Reason being is that broad match requires a smart bidding strategy to actually work well. So when you're using a non-smart bidding strategy on broad match doesn't work. Number four is display being enabled in

search campaigns. this is a big rookie error and you're just wasting budget on the display network which is arguably one of the worst advertising networks that exist like you don't want spend going there particularly on cold. The next one is no negative keywords. This applies to no branded negative keywords on cold campaigns but also no negative keywords in general on your search campaigns as well. The next one is the auto apply recommendations are left on on the account. So, Google is just randomly going and applying stuff into the account that almost always is not in

your best interest. So, they're the main mistakes to look out for in your account. You're probably making one of these mistakes right now. All right, so let's talk about performance max. We're going to cover the mechanics, the traps that you'll fall into, rules for segmenting asset Groups, brand exclusion lists, search themes, audience signals, and the conditions in which we have performance max in a campaign versus when we don't. It's worth noting that Pmax is not the enemy. If you go back really far on the BlueSense YouTube channel, you'll find a video that I put out

called something like Pmax campaigns are terrible. And that was a bit of a biased view that I had at the time, but it was true and it was contextual to the moment in which that video was made, which is that we were testing Pmax a lot on accounts. And what we were finding is that they weren't as incremental as shopping campaigns when you read the rorowaz number. And so if you went into an account and you saw Pmax was at a 5x and let's say you saw that shopping sitting next to it was also at

a 5x. The test that we were running at the time was what happens if we go and scale this Pmax campaign by 2x over the course of a 30-day period. So we ramp it up. And then on another account, what happens if we ramp up the standard shopping? And what we saw every single time we ran the test was that the incremental revenue return on the back end of the website was way better on the standard shopping ramp. And so we at the time put ourselves into this perspective and view that Pmax is just a

bad campaign type. It just doesn't scale anywhere near as well on cold audiences despite what the inplatform numbers say. Because the caveat of this whole thing is that when you go and scale the PMAX campaign, the row still says five. You go and scale the standard shopping, it drops. It goes down to like a four and a three. But the actual back-end revenue is better in this circumstance. And the reason why that occurs is that Pmax because it retargets a lot, it ends up overcrediting just for more purchases as you put more and more spend

through it rather than truly driving new customer acquisition in the business. Whereas the standard shopping campaign drives very Top ofunnel traffic that then might not get attributed to standard shopping. It might go and get attributed into PMAX because it goes and retargets, follows up and gets the final click before the conversion. And so this is really the dichotomy of what Pmax was like within the first 1 to two years of launch. But I will say that Pmax isn't the enemy. PMAX with the default settings that it has when you launch it for sure is. But

if you set PMAX up with the correct asset group structure, you have brand exclusions in place, you understand that audience signals are used for signals, not for targeting, and you use the feed only strategy where it's appropriate in the business, then it's a perfectly good tool to have within the structure. However, caveats are Pmax is a great tool for smaller accounts. If you're spending like 15 to 20k a month or let's say even less than this, Pmax is good. It'll get the account moving. It'll get it scaling to this level. But quite honestly, Pmax does

hit a ceiling. You scale and scale and scale it and then eventually you can't push it past a certain point. At least you can't push it past a point where incremental returns on the back end of the website actually follow what the platform's telling you. And the real big issue here is that once PMAX does plateau, you go and scale and scale it up and then you find this point in which you can't scale it any further. The natural response is either well let's just raise budgets even further and see what happens and just let

it run for a longer period or let's raise the target return on ad spend on the campaign to try to get it to operate at a higher efficiency here so then we can afford to put more budget in. But the issue is both of those moves just pushes more spend into warmth and repeat traffic and the PMX rows might go up but it just becomes a glorified retargeting campaign. I would start to Reframe the question on Pmax away from should I run performance max campaigns to instead what percentage allocation should I have to Pax versus

standard shopping because the answer for most mid to upper market ecom and retail businesses is not a lot to Pmax 20 30 40% maybe but a lot of spend should be driven directly through standard shopping and if you're a lead genen business watching this every time I say shopping just imagine I'm talking about search so the single biggest mistake that's made in Pmax is just running one asset group when you go and set up the campaign It just by default creates one asset group. And so if you're lazy, you just click through and you don't

actually set up multiple. An even bigger mistake is that people will set up multiple asset groups, but they'll forget to customize the listing group. And so you just have all the products bundled into every asset group anyway, and it doesn't even matter. So what do these actually mean? If you've never even opened up Google Ads before, or you're a founder trying to understand this better, so you can have more in-depth conversations with your agency. The asset group is effectively the same thing as different ads in Meta or different ads within a search or shopping campaign.

When we talk about shopping campaigns though, we technically don't have different ads. We have different products that come through from the feed. And so the product feed within Pmax is called listing groups. And so this is effectively how we are bundling all of the products that are flowing through from the GMC feed. With each asset group, we have a connected listing group. So the asset group has all of the assets, so the images, the headlines, The descriptions, etc. All of this information provided in the asset group allows the campaign to go and take the headlines,

titles, descriptions and serve onto search, serve onto display, serve onto YouTube as the little side banners, serve on to discover, serve onto Gmail, and any other random placements that it wants to go and put you on, it will use these assets to do so. So, this constitutes a large portion of what your ads actually look like. But if you're an e-commerce brand and you were listening earlier, majority of your performance should be coming from shopping. And so to get placements onto shopping, it uses the listing groups and just places these over onto shopping. Now, it's

really important that when you build out multiple asset groups, you segment the listing group accordingly based on how it's structured. Now, the question you should be asking is, wait, why are we segmenting at an asset group level when the number one rule of account structure on every single platform is consolidation beats segmentation? Why are we segmenting? And there's four core reasons. Number one is we want creative relevance. So if I go to the website and I look at jeans, when you start retargeting me across display, across all these different channels that Pmax can place on,

you want to retarget me with photos and headlines of jeans. You don't want to retarget me with random other stuff that's not related to what I was actually looking at. Now, if you just group everything together into one asset group, I will just get served random stuff. But if you segmented out and you had a jeans asset group and then you just had the jeans products from the listing group in there, then when I go And look at jeans, it will dynamically retarget me with relevant assets. This is also the case on cold targeting. So

even though I don't love it, I don't love this going out and just targeting cold on display and YouTube, etc. But when it does inherently go and do that, it will go out and if it's all grouped together, it will just go out and serve all of your fashion products to anyone that it thinks is interested in the brand. Instead, if you have segmentation, it will go out and take your jeans products and serve it to people that it believes are interested in market right now for buying jeans. Much more contextually relevant, much more relevant

on a creative angle and therefore you see better performance. Number two, which I kind of hinted towards a little bit there, is audience signals. So on each of these asset groups when you set them up, you have to also select an audience signal. Now, the big mistake here is people think audience signals is like targeting in meta ads back a couple years ago where you choose your interests and then it goes out and targets those interests. That's not how it works at all. Audience signals are just directional signal that the campaign uses at the start

before it has a lot of conversion data. So, when you first launch the campaign, you just got to give it a little bit of direction rather than it going off and just randomly testing. Google tries to not waste your budget in doing so. And so, it says, "What kind of people should we be targeting?" And so you go well for this asset group you should be targeting people that are interested in jeans in market for denim uh that are this age that look like this that are interested in this stuff. And so then out of

the gates rather than wasting budget on testing random people it goes straight to directionally the people that are Interested. Now obviously audience signals work better when you're more granular and you have that segmentation at an asset group level because if it's all consolidated you would just go people in fashion but if it's segmented you can go no no no no people that are in market for jeans right now and are making searches for gene related products. Number three is search themes. This is effectively the same thing as audience signals, but search themes is saying, "Hey,

if they've searched for keywords like this, then target them." And then number four is visibility into performance breakdowns. So when you introduce segmentation at the asset group level, you get the ability to actually see performance on each asset. So rather than it all being bundled together, we can go, "Oh, wait a second. Tops are currently outperforming jeans. That makes sense because of the seasonality that we've checked against the keyword planner and the demand forecasting of the business. All right, let's think about how we can redistribute budgets accordingly or make changes based on this data insight.

If you don't have asset group segmentation, you don't get that insight into the data siloed in that way. So what are some general rule of thumbs for segmentation here? Number one is you generally want one asset group per product type. So in furniture, if you have chairs, tables, lighting, decor, you want that split out at an asset group level. In for example hardware you would want hand tools, power tools, fasteners, hardware accessories split all out into different asset groups. Inside each asset group, really important that you customize the listing group accordingly so that if you

are just putting hardware accessories in here, the actual products in there are just hardware accessories. So you want to make sure that that segmentation is set up. Number three, you really want and this is going to depend on actual Skew count in the business. So take this with some flexibility, but you want at least 15 products in an asset group or else you are really just oversegmenting. What you definitely don't want like a real red flag here is you don't want one product in an asset group. And the reason being is it just won't get

any spend. And so if you have a bunch of asset groups and then you're throwing in a couple asset groups, like let's say you have a top performing hero product and you're like, I want this in its own asset group. If there's only one product, it might actually get not as much spend as it should. And you obviously want to take into consideration the exact same thing that you take into consideration in account structure, which is that consolidation will always beat segmentation. When PMAX first rolled out, we did tons of testing on asset groups. We

had accounts where we were launching like 200 asset groups with this hyperderee of segmentation. Different asset groups for different listing groups, different search themes, different product types, like the list goes on. What we ended up finding was that the simpler structures always performed better. And so you want segmentation. You want it out by different product types, etc. But don't go and throw 400 asset groups in the account. I'd even say like as a general rule, have minimum three. Have maximum 20. When you start to go over 20, I don't see many businesses that have enough

complexity to be able to warrant that degree of segmentation. So we then move into search themes. How do you select your search themes? How important is this? How does it work? There's three things to know. Number one is that search themes function as inputs to Pmax, but they are signals and they're not targeting. So when you go and put in a bunch of keywords or a bunch of search themes that you think are relevant, it's not guaranteeing that the PMAX goes in places and bids on those key terms. Once again, it's just directional signal. Number

two is that when you set these up, you want to prioritize the key terms that are actually converting in the account. So you want to go have a look at your search campaigns that you've run historically, what key terms converted well, and use them. Don't make stuff up here. If you can use data from historical performance in the account, that's always going to be better. And then, as I said before, and this might be a little bit controversial, but honestly, this is what the data shows us. After running PMAX on over 250 accounts, consulting on

another 400 ecom accounts from 7 8 9 10 figures is that search themes and audience signals don't really matter if you're a mature account. Kind of the reality. If you're already spending 30, $40, $50,000, $100,000 a month on payax, your audience signals, your search themes, they're not really going to move the needle. It's like it's a 1enter at best. And so, if you get an audit from an agency that's going in and going, "Oh, your audience signals can be tweaked a little bit to look like this, or your search themes could look like this." It's

going to make literally no impact to the business. Where search themes actually matter is on a new launch. And so this is either literally a new launch of a new business and this is the first campaign that you're launching or new launch of a new product or a new asset group or a new category. Okay, if there's no historical data, that's where the PMAX then goes and looks at the search themes and the audience signals and goes, okay, what is the user telling us? What direction should we go in? And then it starts there. And

then the same thing applies for audience signals as well. So, Google will just bypass the audience signals once it has real conversion data to work off. Where they're actually helpful is when you're warming up a new campaign or a new asset group, when you have a new account and the algorithm has no signal. But after 30 days of conversion data pulling up within the campaign, the audience signals done its job and you could remove it from the campaign and it won't impact performance at all. Now, when it comes to brand exclusions on Pmax, this is

relatively straightforward. Simply exclude your brand name from your PMAX campaigns. There's actually two ways to do this right now. Now, if you go into campaign settings and then you scroll down and then you click brand exclusions in there, you can then click add new brand. You need to put your brand in. Google will go and index it, find all relevant branded key terms and then you can select it as a list. That's called applying a brand list exclusion to PMAX. This was the only way in which you could actually exclude your brand for the last

like year. Then recently, Google went and rolled out negative keywords on Pmax, which is amazing. So now you can go in and you could just negative keyword your brand directly in here. Now what I like to do is both. Let's put fail safes in place so that if we miss a negative keyword variation, it doesn't go and place on it because the brand list scoops it up. And then vice versa, if the negative keywords stop working for whatever reason on the campaign, at least we got the brand list there to actually protect us. If I

was to go through and order the importance of all of the variables that matter in Pmax, number one is the actual GMC feed. It is the titles. It is the descriptions. It is the imagery within the feed. Assuming that this is e-commerce and the Pmax is putting majority of spend into shopping, right? So, I'm going to put an asterisk here because this is contingent on the PMAX actually prioritizing shopping and this obviously being an e-commerce brand. Number two is brand exclusions. If you don't have brand exclusions on your PMAS campaigns, they are just retargeting campaigns.

Number three is asset and listing group structure as Well as obviously the assets in the asset groups. If you have terrible headlines, terrible images, terrible descriptions, the ads are going to look terrible and therefore they're not going to work. At the end of the day, the creative of how the ad shows up is vitally important. And so the assets and the structure and the copyrightiting etc in there is going to be third. Number four is the bidding strategy, which we'll talk about bidding strategies in more detail later on. And then number five is the audience

signals and the searchs. So before we move on from PAX, there is one variation of Pmax that exists and you've probably heard it before if you're deep in the Google ad space. If you're not, if you've never even opened a Google ad account before, it's going to be new to you. This is a feed only PMAX. Now, how this works is those asset groups that we were talking about before, you delete them. And when you delete the asset groups, all that's left is the listing group. And the listing group is the products flowing through from

the Google Merchant Center feed. And so what happens is PMAX can't place anywhere except on shopping and except on display and it will only do display retargeting because it can't pull through headlines or descriptions or anything. All it can show on display is the GMC feed. So the product tiles popping up. This is effectively what smart shopping campaigns were back in the day. And for those that haven't been around long enough on Google ads, smart shopping is what Pmax used to be where it would only place on shopping and do display remarketing. Now, a little

tip here for feed only campaigns is there's two things that end up destroying them, which is that number one, URL expansion being on within settings. You need to make sure URL expansion is off or it just goes and starts adding URLs, scraping information, building asset groups automatically without you doing It. And then number two is having automatically created assets on within settings as well. So you need to make sure those two settings are off if you're going to run feed only. Now why would you run feed only? It's so that you can force spend into

the shopping placement. If you don't want pmax placing on search, YouTube, anything else, you're like, we just want Pmax placing on shopping. That is it, then this is the effective hack around being able to do so. There's a lot of reasons as to why you might want to pivot into feed only. There's a lot of reasons as to why you might want to run feed only on an account. I'm not going to go into all the reasons here as this is a little bit more of a complex nuance strategy that's going to be context dependent

on the account. But let me give you one example, which is that we had a client uh spending a couple hundred,000 on PMAX per month and then the issue was is that clicks went through the roof out of nowhere. Return on ad spend dipped by about 20%. And we what's going on? Is this bot traffic? What's actually occurring here? And what was happening is the PMAX campaign went rogue and started distributing about 10% of spend into YouTube where it was previously only distributing 2%. and it went and started distributing like 10% of spend into Gmail.

Gmail cold, which is crazy. And so, as a product of that, click volume looked really high and good. The clicks actually weren't translating to the website. Here's a little side note is that Google PMAX campaigns will place on Gmail, but when they place on Gmail, and you'll see it is just like a fake email at the top and you click on it and it just shows you an email that it puts together using the assets and the GMC feed is that the click onto the email, me just opening the email up, that's what the platform

counts as a click, not an outbound click to the website. And so When you place on Gmail, you get a ton of clicks that seem to be like 10-centent clicks, but they're not actually clicks. These people aren't actually going to your website. They're just opening the email that was in their inbox. And then what you also end up with is a lot of customers getting really angry saying, "We have unsubscribed. Why are these emails popping up in my inbox?" And then you have to say, "Well, they're not actually an email. They're an ad." And then

the customers don't believe you. So it's a bit of a nightmare all around. So in that circumstance, what did we do? We pivoted the PMAX out into a feed only so it couldn't physically place on Gmail. So it couldn't physically go and overspend on the YouTube channel. And because of that, we fixed the issue and resolved it and that account continued to scale. And so this is really an ability to crutch and solve problems if problems arise with the primary PMAX campaign. You might also decide to have multiple PMAX campaigns for different categories and then

some categories are feed only, some categories aren't. The last comment I'll make on Pmax is T roaz. So most people's understanding of bidding strategies is fairly rudimentary. And so you have target return on ad spend. The other option here is that you have max conversion value. Now what people will do is they will launch on max conversion value to accumulate data and then once data has accumulated they will roll into a target row strategy. And this is how the platform works and this is generally what you should do. Reason being is that when you're using

max conversion value, this is what performance looks like over time. When you switch to target rorowaz, this is what performance looks like over time. It becomes a lot more stable. So The campaign rather than making big bets that are going to pay off and you're going to see better efficiency and then it makes big bets and you're going to lose and you have this instability week on week, target rorowaz will flatten it out. Now notice the actual performance over this whole time period is the same. You don't necessarily get better performance with target rorowaz against

maximized conversion value. Performance ends up averaging to somewhat similar. The only core difference is the reliability of the campaign on a day-to-day basis. Now, most people want better reliability. They want consistent revenue coming from the platform every single day. And therefore, target rorowaz is a good feature to roll into. The trap people fall into is that they roll into target rorowaz. They maybe set it 400% because their trailing return on ad spend is maybe 420 430. So they set it a little bit under and that's fine. That's all well and good. And the campaign starts

to stabilize and it hits that rorowaz. It's a lot more stable on a day-to-day basis and we go fantastic. We don't want to scale. We don't have the budget to maybe we're agency side and the client is strict on financial year budgets that the board is given. And so as a product of that, what can we do? Well, we can increase the target rorowaz to try to squeeze more efficiency and more revenue out of the existing campaign. So you'll come in and you'll go, "Okay, let's bump this to 425." And then the rorowz goes up

a little bit. And then you go, "Okay, cool. Let's bump this again to 450." And then the rorowes goes up a little bit. And then you keep doing this and maybe you end up at about 500%. At this point, spend starts to pull back. It can't achieve that rorowaz until you find Equilibrium and you find, okay, this is the max we can go to. Now we're at 500%. Zoom out, look back, and you go, this was incredible media buying. Well done to the agency. Well done to whoever did this because at the start of the

quarter we were getting a four rorowaz on this campaign spending $1,000 a day. Now we're getting a five rorowaz and sorry this is cuz we're looking in percentages. This is a five rorowaz. This is a six rorowaz. Um now we're at a six rorowaz at $1,000 a day. Great. Reality is this down here was probably a better position for the campaign. And the reason for that is because this was likely going after more cold traffic because the target return on ad spend trap is that as you increase target rorowaz for the campaign to achieve this

return what it does is it restricts who it bids on and what it restricts is cold audiences. If you think about the entire population in here and then you think this is the people that I want to target that I can achieve a particular efficiency on. Now this target demographic is probably going to have a lot of warm people in it is the reality. There's probably going to be people that have already visited the website before probably some existing customers probably some people that have seen your ads on Facebook. And if you just go and

target this narrow amount of people you will be able to achieve a 7x return. But then if we want to scale spend, what happens is that this opens up and we target more cold users. And as we go out and out and out to colder and colder users, we get worse efficiency. The Opposite then occurs when we increase target row. So if we go and increase it, let's say this circle is a seven and we want to go and take it to an eight, what happens is the circle shrinks and we actually go in and

just start targeting this area of people. And so even though efficiency might look better, even though we're happy, we're actually not genuinely acquiring new customers and we're just tightening the pool of users that we're targeting. This is where you start to get into feeder strategies. The core premise of a feeder strategy is that you have a campaign with a really low target rorowaz. It could be as low as, let's say, 50%. So that's a 0.5 rorowaz. And then the idea is that when we go back to this pool, this campaign is going to target everyone

in the entire pool. It's going to reach everyone that's searching for these key terms. And then you have a second campaign which has a high target rorowaz which might be like let's say 600%. And this campaign will only target warm audiences. It will only target people that are in market and highly likely to buy and have probably visited the website before. Idea is that this campaign drives all the traffic captures as much as possible and then this campaign converts [clears throat] them. So you're effectively feeding traffic from the top of funnelunnel acquisition campaign into the

bottom of funnel campaign. The added advantage here is that because most people don't operate a campaign on such a low target rorowaz is that you're going to reach people that no one else is even entering auctions on. And so where this might be the total pool of users that are searching for relevant key terms to your business. Let's say it's fashion and let's say it's dresses. So this is everyone searching for dresses on a day-to-day basis. Most people are only bidding on this half of users. And the reason being is that these half of users

have really high intent, have searched for a lot of purchase relevant key terms recently, have visited a lot of websites. And so Google knows these people are very likely to buy. And so anyone that has a high target rorowaz strategy is going to go and bid here. And everyone's going to be bidding against each other. But all of these users from Google's perspective look super low intent. They haven't visited any websites before. or this is their first time searching for this kind of key term. And so because of that, almost no one is out here

bidding on all of these auctions for all of these users. And so the idea, the arbitrage opportunity is that you can just go and bid on all these users. It's super cheap to do so. You can drive tons of traffic over. Now, yes, a lot of it might not be that good, but a lot of it might be good, and we can therefore funnel it in and convert it on second or third click using the bottom of funnel PMAX campaign. And this actually ends up being a really effective strategy that we see work on a

lot of accounts. The reason why a lot of people adopt this strategy is because PMAX was sold to the community as a replacement for smart shopping back in 20122. But 3 years later, we're sitting here and we're measuring all the results. We're looking at the findings. I'm talking to other agency owners who own some of the biggest Google Ads agencies in the world. And everyone pretty much says the same thing behind closed doors, which is that Pmax substantially overattributes in the platform. 50% or more of Pmax conversions are repeat or warm customers that Pmax didn't

actually genuinely acquire. Standard shopping now Substantially underattributes cuz Pmac takes Pmax takes all the credit for it when they sit next to each other. And so the combined effect is that Pmax looks like it's scaling well in an account, but it's actually being held up by standard shopping doing all of the cold acquisition work. Now, generally when you're running a feeder strategy, the actual approach here is that you run standard shopping as the cold campaign because it's just better at actually placing and reaching cold audiences and then you run PMAX as effectively the high Tores

retargeting campaign. All right, so here's three different strategies that are in market that a lot of people use that I like that you can do with standard shopping. Now, the same rules as Pmax and everything else that we've talked about still applies here. Okay, you want consolidation over segmentation. If you are going to segment, it should be based on those four variables that I mentioned earlier. Um, but here are some strategies in market that seem to work decently well. Number one's pretty straightforward, which is you just have all your top sellers in one standard shopping

campaign and then you have everything else in another campaign. The idea here is that you want to force into Parto's principle and have 80% of your spend in the top 20% of products. Um, but it also allows you the flexibility to continue to ensure testing budget down here so that you can roll new winners up. This is particularly just relevant if you have a really large skew count and you're not fashion. If you're in fashion, I probably wouldn't do this. And there's a lot of reasons as to why. And if you're 5 10 product business,

you also shouldn't do this. You don't have enough products to really warrant this degree of segmentation. So, this is a Very specific case for a business that has a lot of products that isn't in fashion. This is a structure that works. We then have the feeder strategy, which I talked about before, which is that you have standard shopping then feeding into a high trow pass campaign. And then you also have the flow boost labelizer. It couldn't be a Google Ads video if I didn't mention this. Uh because this is very common on any people that

are deep into Google Ads. What this really is is just splitting up your ad groups or your campaign into different performance levels of products. So you have overindex index, near index, under index, no index, which this means that it's above expected performance, it's hitting target, it's just below target, it's below target, or there's zero performance at all. Now, if you don't have a lot of products, you could just do this manually, and you can just segment the products out and give them custom labels. But what you can do instead is you can set up an

automated uh labelizer script, which will just autotier them on 30-day rolling performance in real time. So, effectively, your products are going to get shuffled around at an ad group or a campaign level based on their real-time performance, and then they will have a bidding strategy and a budget that is relevant to how hard you want to push the product in that given state. Once again, I don't recommend this to majority of businesses because it introduces a lot of segmentation for not a huge amount of upside unless you have like thousands of SKs. When you have

thousands of SKs, this becomes a really effective way to manage it and be able to stay on top of it. But for most people, this is just over complicating Google Ads and you don't need to go this far. Then we've got search campaigns. Transparently, search campaigns is not something that we focus on that much. It's because of what I said before, which is that for 90% of e-commerce brands, 90% of your focus should be going into shopping because that's where the greatest leverage is going to be. Now, that breaks in one particular circumstance in ecom,

which is in ecom businesses that have a B2B component. So, if you sell to businesses in some capacity, but it is through an ecom store, you do actually see arguably better performance on search campaigns over shopping. I'll give you an example of this which is that if you're selling bulk eyelashes to like technicians and small businesses in that case you end up with better performance on search and the reason for this is because when you are making a business search you often don't click on a shopping listing because it doesn't match the price discrepancy that

you have in your head. So, if I'm going to go and make a business purchase, typically I want something in bulk, right? If I'm going to buy like tea for my coffee shop or my restaurant, I either have a known supplier or I'm going to go to Google and try to find a supplier. Now, when you're going and finding a supplier and you see a bunch of shopping listings that say tea is $5, $10. I'm not going to click on them because I know that do these people actually do bulk rates, do they do business

deals, you just know that B2B business does not take place on the shopping network. it's intuitive to the user and so because of that you go to search listings and you start clicking on search ads and so that is the case where B2B actually makes search campaigns more important. So there's a few different components of search campaigns. Number one what we talked about earlier which is you want to be using topic ad groups not keyword ad groups on negative keywords. The first 30 days of a search campaign is the Highest leverage point for negative keywords

because the campaign's going to go out very broad and target a bunch of keywords and there's going to be a lot of stuff in there that is actually irrelevant that you don't want to place on. And so as a product of that in the first 30 days I would be reviewing negative keywords weekly dependent on budgets. If budgets are really high this can literally go down to a daily level so that you're not allowing any kind of keyword bleed and waste money. From month one onwards you can drop this to weekly-nightly. You actually don't want

to tweak with negative keywords too much as every time you do it, it does reset learnings and impact performance a little bit. On top of this, you also want a universal negative keyword list that's just applied at the account level. So, every single new campaign inherits it. We have one of these internally. You can literally just go and search up for one. I think they'll be all over the internet or you could ask an LLM to try to build one for you. It just has stuff like job related queries. So if a key term has

jobs or careers or interns or hiring or salary stuff like that, you just don't want to place on that key term because it's not relevant. And just a quick note on responsive search ads, which is the only thing that you can run these days, is reiterating what I said before. There is 2,700 possible combinations of a search ad when you set it up. If we assume it takes 100 impressions to be able to determine whether a specific combination works, which is honestly really low, like I wouldn't want 100 impressions of data, but if you assume

that, then it takes 270,000 impressions to be able to actually exit learning and have the campaign understand which combo works. And so the implication of that is that you want to Minimize segmentation because the more segmentation you have, the longer these RSAs will be in learning. And in most accounts, they're just in learning indefinitely. One final note here that isn't spoken enough and I actually think it's one of the largest levers that exist within search campaigns is nothing to do with Google ads. It's not even in the account. It is the landing page. And the

reason being is that if you think through search campaigns and Google ads more as a whole, if we just step back out of the platform and we get out of the weeds and all the tactics and what we're actually doing here and we just think through spending to acquire customers, what we are looking at is that we are charged a cost per click on Google and then we are driving this to the website and then we are getting two things out the back. Ultimately, we're getting revenue, but revenue is a function of the conversion rate

on the website and the average order value of those conversions that are occurring, that ultimately will then equal rev. Now, when it comes to CPCs, there is a little bit more nuance within this, which is that if we just arbitrarily optimize for the lowest CPC, it doesn't necessarily improve this equation. In fact, CPCs are actually a vanity metric, which seems counterintuitive because you look at this and you go, "Well, all that actually matters is we need the highest average order value, the highest conversion rate, and the lowest CPCs." Not true. And that's where people get

into trouble. CPCs, and you see this on any large data set and analysis, CPCs and ROI, and you can measure this on acquisition, me on just rorowaz in the platform, doesn't matter what it is, there is very little relationship whatsoever. You look at accounts and some accounts have very high CPCs with a Very high rorowaz and vice versa. You see this on a product level, you see this on a campaign level. On the peripherals, it's true. If CPCs are like $100, yeah, you'll never be profitable. And then vice versa, if CPCs are like 10 cents,

it's probably junk traffic and you'll never be profitable. But once we tighten into the middle, CPCs don't really give us any good indication of performance. And that is because the quality of this traffic changes. You can have high quality clicks, you can have low quality clicks. And the quality of the clicks is ultimately based on the automated bidding that is occurring in the platform. And you are trusting that the campaign is going to bid correctly on users that are likely to convert from you. Which means that all of Google ultimately just comes down to the

auction that you have against competitors. You and all of your competitors are bidding on a specific key term to rank. And at the end of the day, whoever can pay the most amount of money to win the auction wins the listing, gets the click, gets the conversion. Now, if Google is just all about who can spend the most money at auction to get the placement and get the conversion, then how do we make sure that we can pay the most amount of money? Well, it is by having the highest conversion rate possible on the website.

If we are converting 2x higher than all our competitors, well, we can spend double the amount on a click and we'll make the same amount of money as them. And that's how we obviously flood Google and beat them. Same thing for average order value. If our average order value, then we squeeze more revenue out of the same amount of clicks and therefore we get an advantage and we can spend more. And so ultimately all of the stuff that we're doing in the platform, the segmentation, the asset groups, the audience signals, the optimization of The GMC

feed, which we're about to go into, all of these things are trying to improve the quality, which is improving the auctions that we enter. So we enter better auctions. we don't waste on auctions that don't matter on lowquality users and it's ultimately improving our quality score on the platform so that our bid that goes into the auction is lower artificially than competitors because Google's giving us an advantage and so that's ultimately everything we're trying to do in the platform is here but at the end of the day with all of that in place we're still

going to be restricted based on the fact that if a competitor has higher conversion rates and higher average order value they can just still out bid us with worse ads with a worse campaign structure with worse everything and and they will still beat us because they can afford to. And so when it comes to search campaigns, what very little people are doing is just split testing landing pages. It's just that simple is that when you have an ad group or a campaign set up, rather than just having one ad, have two, duplicate it, have two

ads, but ad one goes to one landing page, landing page one, and this goes to landing page two. And then we monitor conversion rates over a 30 to 60 day period. And then the landing page with a better conversion rate, that gets selected. And then what do we do? We rotate in another test and we have add three. This gets turned off and we roll in. And let's say that this landing page was even better. Amazing. And then slowly we're increasing the average conversion rate of the campaigns which is allowing us to bid more. Ideally,

if we can do some average order value optimization too, even better. And so our revenue per visitor from the campaign improves, which gives us the ability to unlock scam. I strongly Believe that landing page testing on search campaigns is actually one of the biggest levers that you have. Before we move into bidding, two quick side notes. Make sure you're excluding brand on search campaigns. And then when it comes to bidding strategies, there's so many different tactics here, but generally speaking, I just go with a smart bidding strategy like target rorowaz or max conversions or something

similar. Let's talk about bidding strategies at an entire account level. So, what are your different options? Which one should you go through? How do you understand the differences? Let's start with the ones that I really don't like because we only work with established businesses. But if there's people watching this that aren't established, I have to mention it, which is that if you have a new account with no conversion data, you actually won't even be able to use smart bidding strategies, at least the last time I checked. And so, you'll be forced into using a manual

bidding strategy. You should use manual CPC or maximize click. If you are an established business, you should not be using either of these strategies unless the agency is very, very competent and has some kind of complex intricate strategy as to why they are applying this. We will sometimes do some clever tricks and strategies in an account to try to squeeze out more performance. And it might involve using something like maximize clicks. It's rare. You won't see it in 90% of accounts that we manage, but it's a test that will sometimes run to see if we

can get a squeeze or take advantage of an arbitrage opportunity. This really should just be beginners and ideally you get off these bidding strategies as quickly as humanly possible. Then you roll into maximize conversions or maximize conversion value. These are what's called expansive bidding strategies. And the reason why it's called expansive and target rorowaz and target CPA are restrictive is because Target CPA and target rorowaz restrict who you actually enter the auction on based on an efficiency target. You're putting a limit on it and saying don't enter auctions unless you can guarantee this particular return.

But on expansive bidding strategies, they can go and enter whatever auction they want. Now obviously it's going to do it within uh the constraints of actually getting you results but it will be more aggressive and it will go into auctions that otherwise you wouldn't have gone into if there was a restriction on the campaign. Now there are benefits and disadvantages of that. The benefits are you will win auctions that you otherwise wouldn't have won. The disadvantages is you will lose auctions that you otherwise wouldn't have lost. And so this is effectively a higher risk model

to run on. Now a core thing to understand here, easy way to conceptualize the difference between expansive and restrictive bidding strategies is percentage allocation of budget towards testing. So if we were to just theorize this for a second so you can understand the concept, let's say that maximize conversion value campaigns, they put about 30% of budget towards testing. And what I mean by testing is that it's testing new keywords, new audiences, new if it's performance max, new placement types, new regions where people haven't bought from before, more new psychographic data points. It is testing stuff

that based on historical conversion data doesn't seem like the best thing to test. It isn't it hasn't converted before, but it's adjacent. It's tangential. It seems like ah it's a worthy test. Let's put it in and see what happens. And so because of that, this campaign continues to test new Audiences and find new audiences that work, which unlocks scale. And so it will go and test on all of these different people and it go actually people in this region in this target demographic are actually learn working quite well. Let's double down, spend some more money

there. And that's what allows you to unlock scale and continue to increase budgets. Whereas with a restrictive bidding strategy, there's about a 0% testing budget. Now, this percentage allocation really is contextual to how aggressive the target rorowaz is against the actual rorowaz in the campaign. If you're achieving a four rorowaz and your target rorowaz is a 4.5, you'll definitely have 0% testing. It's going to put all the budget into ensuring that it can actually drive the efficiency you want. If instead there's a big discrepancy, so maybe you're hitting a five rorowaz, but your target row

is on the campaigns at two, well then you actually might have some testing budget in there and it's going to operate more like max conversion b. Now, the advantage here is that you don't have any testing budget, so you're going to be more efficient. Uh the disadvantage is that you're going to be siloed into a particular audience and a particular target demographic and it's not going to evolve over time as the business evolves and as the platform evolves. And so just being on target row permanently for an extended period of time can be a dangerous

position to be in. Ideally, we like to rotate between the two over extended time periods so we don't pull the account into a position where it's too siloed in in terms of data and it starts to negative feedback loop. But we also don't want to just be on maximize conversion value all day and not be as efficient as we could be. Now the question also becomes now that you understand the difference between expansive and restrictive well which of these should we use? Should we use maximize conversions or should we use maximize conversion value? And then

that will obviously also change which version of these bidding strategies you roll into because TCPA is a version or a Subcomponent of max conversions. Tores is a sub component of max conversion value. Now in e-commerce I would generally 90% of the time recommend that you just go for value. And the reason being is that when you go for conversions is ultimately optimizing for the lowest CPA possible. And the issue with optimizing for the lowest CPA or the lowest CAC cost to acquire a customer, it will optimize towards pushing your cheapest product because if you have

a, let's say, a $40 t-shirt, it's much easier to sell a $40 t-shirt than it is to sell a $200 jumper. And so as a product of that, all the spend just ends up going here because you could probably achieve like $20 CPAs on this t-shirt. But on this jumper, it might cost you $60 to be able to actually acquire an order. Now, the rorowaz here might be very similar. It might be the same, but because this has a lower CPA against the spend. And in fact, in this instance, you can see the rorowz here

is actually better. We would prefer budget goes into the jumper. We're getting like a 3.5x ROI here. 3.5x. Over here, we're going to get 2x. But campaign doesn't care. It's not optimizing for rows. It's not optimizing for value. It's optimizing for the lowest CPA and the most amount of conversions, most amount of orders. And so, as a product of that, all the budget goes here, which is not what we want. Hence, max conversions can actually, and it typically does end up putting a lot of your spend into products that you don't want to put money

into. So, a few more quick notes on bidding is that you will see a target rorowaz spiral. Um this is super common in Google ad accounts which is that when you look at the account over time spend will be like here and then it will just slowly trail off and campaigns will just Die. And the reason for this is that if we overlay rorowaz rorowaz might have dipped and then because the target rorowaz is sitting somewhere like here the account is no longer hitting the rorowaz goal and so spend pulls back. And what this really

is is the bullseye analogy that I kind of gave before, which is that you are targeting these people and then as you stop achieving the target rorowaz, you need to shrink the circle even further to make sure you still achieve it. And so the circle gets smaller and smaller and therefore spend starts to fall off. So once you get into this circumstance, if you leave target rorowaz where it is, entire campaign declines and just dies. If you start pulling target rorowaz back, you need to pull it back aggressively or else it continues to die. What

people do is they're not aggressive enough in pulling the troz back. So the entire account just continues to decline. Now the reason why this spiral continues is because as you are not achieving the target rorowaz, conversion data falls off and target rorowaz uses primarily the last 30 to 90 days of conversion data to be able to model who it should target. Now, as your conversion volume starts to decline, the modeling accuracy starts to decline and this is where you end up in these negative feedback loops because let's say in the last 30 days you were

getting 100 conversions and it was using those 100 conversions to optimize and figure out who to target and then now this suddenly drops for whatever reason. Maybe it's seasonality, maybe a product went out of stock that was good, something happened, your conversions drop to 80. Well, now you have less conversion data to be able to model off and because of that you have Small sample size bias. Therefore, the targeting gets worse. Therefore, the conversion volume falls further. And as the conversion volume falls further, you stop hitting the target rorowaz. And therefore, the spend starts to

pull back. And as the spend pulls back, you get even lower volume. And then even lower volume. And as the volume declines, the accuracy of the bidding declines and the bidding model starts performing worse. And you just negative death spiral. This is what you need to be really cognizant on and careful because if you don't know mechanically why this happens, how it exists. You see campaigns all the time that just do this. We on board campaigns that are like here a lot of the time and I say in the audit I'm like this is exactly

what's happening right now. You need to fix it soon or you're going to end up here pretty soon. And these Google campaigns aren't going to be driving any volume for you anymore. Also, a specific hack worth knowing is that there's stuff called portfolio bidding strategies. And this is where you set the bidding strategy up at the account level. Now, the reason why you would do this is it gives you access to an extra feature which is kind of cool. You also get access to this feature in SA3, but that's only for enterprise businesses, so it

doesn't apply to most people. Um, and portfolio bidding strategies actually allow you to merge learnings, which is really cool and really critical as an implementation if you have too much segmentation in the account. What am I talking about? Right at the start of the video, if you remember, I drew out campaign ABC and I spoke about how these campaigns don't share learnings. They're isolated. Hence, segmentation is not good. You can actually fix this. You can take down these silos and allow the Campaigns to share learnings with each other. The way that you do this is

you set up a portfolio bidding strategy that then gets applied to all of the campaigns and the learnings get housed. They all learn together. This is a really cool additional strategy that you can layer in if you want the segmentation but you want the learnings being pulled together. Now the other benefit is that you can use a for example target rorowaz strategy or a maximize value strategy plus implement a max and a min CPC. So if you don't want the campaign for example spending over $10 on a click and let's say that you work in

high-end furniture and so there are sometimes very expensive bids in that niche where you might go out and just bid 10 20 $30 on a user. You're like we don't want them at all. We never want a $30 click. It's not going to be profitable for us. I don't care how high intent they are. You can go up and apply a portfolio bid strategy where you can specify the target rorowaz and then put a max CPC on the campaign. On search campaigns, there's also a setting called smart bidding exploration. I actually really like this as

a concept. Unfortunately, it's not available on Pmax or on shopping. But the idea here is that even with a target rorowaz, it can go outside of the target rorowaz in 10, 20, 30% bands and do testing, which I think is really cool because it's effectively a middle ground between using a restrictive bidding strategy and an expansive bidding strategy where you're kind of getting the best of both worlds. You're restrictive, you're going to hit target rorowaz, but you're like, "Hey, go out sometimes and make some Auction plays that otherwise you normally wouldn't of to be able

to learn some more things and to be able to ultimately unlock scale." So, if you're using search, I'd recommend looking into that setting. Limited by budget. You will see this all over your Google ad account on campaigns saying that you could spend more spend more on these campaigns. Now, this is a product supposedly of the campaign having the ability to enter more auctions where it didn't because of a restriction in budget. So, this is an indicator that you may be able to spend more on the campaigns. This is not a rule that you can spend

more on the campaigns. I have almost always seen on every account, no matter how high we get the spend, everything's always limited by budget. We have campaigns spending5 $10,000 a day on PMAX in a very small total addressable market in Australia, and it says limited by budget. So, you need to be thoughtful about this. Use it as an indicator, but don't use this as the effective bible for being able to determine whether there's incremental spend available within a campaign. Just a quick note on budget changes, uh, which is that the rule that you've always heard,

increase budgets in 20% increments. Uh, it's real. It applies on Google. On Meta, Tik Tok, you have a bit more flexibility. You can increase budgets way faster than that. On Google, I've generally seen it always be a pretty bad idea. When you mess with budgets too heavily, it's way more of a stable platform. It drives conversions way more consistently. And as a product of that, it wants slower, more methodical changes, and you're not just throwing budgets around. Doubling budgets overnight, not a good idea. Having a target CPA overnight, not a good idea. Now, bringing budgets

down quickly, we've generally seen is fine. If you're really high budgets and you just have them, it's generally okay. You don't have much instability. You don't have much of a performance drop off. But going upwards in budgets, you want to move slowly and you want to move slowly in terms of bidding strategy changes as well. We lastly have GMC. Now, I would argue that Google Merchant Center or feed optimization is actually one of the most important levers when it comes to shopping. And the reason being is that how does shopping know what key terms to

place on? It's using the feed. How does Google determine your quality score for where your bids will sit? It depends on your feed. So the most important thing when it comes to GMC optimization is the product title and the product title structure. Most GMC titles are not optimized. This is really underleveraged. How do you actually want to structure the title? You want to structure it something like this. And there isn't a golden rule, but this at least gives an indication as to a title buildout. Let's say that by default you have a dress and normally

dresses in fashion everything has a name. I'm just going to call this the Nathan dress just for the sake of this example. What will by default get pulled into GMC and seren shopping is something like Nathan dress size 12 and then maybe the color will get appended as well. Sometimes you might also have a brand name getting automatically appended to the front or the end of the key term too. So maybe at the start you have the brand here as well. And so that's the default title structure. Now, the issue with this title structure is

that the brand means nothing if we're not ranking on branded keywords because we have brand excluded. So this doesn't mean anything. Also, the brand is at the bottom of the shopping listing anyway. So this like really is redundant and not needed. Nathan means nothing to anyone. Okay, unless people Are product aware and know the exact product that you're selling, this means nothing to anyone. We then have dress, which is the first point in which this actually becomes relevant, but it doesn't tell us anything about what type of dress this is. And then we have the

size, which is also kind of irrelevant, but yeah, it's a nice to have to make sure that people know that it actually has the size that they're in. And so this ultimately just won't perform very well. So instead, what we want to do is take the brand name, push it to the back. If no one even knows who you are, don't have it at all. But if you're a big retail business, still probably have it. Push it to the back. And then at the front we want to start having relevant keywords and context for not

only the user but mainly for the algorithm. So how I would start to reshape this is this can stay if this is the unique name. Okay, we can keep this but let's start to pull some keywords in. And there's really two options here. We can have the keywords right at the start or we can still start with the the product name and type and then go into the keywords. And so in this case we're going and we're adding formal dress to the start which is going to add way more context and allow us to actually

place very well on this key term. Now, why did we put this key term? It's based on actual keyword research of what's performed historically well on the account. So, we're not just making stuff up and putting into the title. The titles are always built based on actual conversion historical data out of search term reports. So, you look at the search term reports, you what is done well. Okay, formal dress performs unbelievably well. This is a formal dress. So, let's append it to the start of this title. Having as much attributes as possible as well as

long as it fits within the title length is also good because this will be used to in placements of key terms. So If someone for example searches for a blue dress, this will then get prioritized in the rankings where previously it wouldn't have because it wasn't actually listed within their title or any of the attributes of the product. And so you want to be thinking through still have the product title or whatever the product's called. Sometimes in other industries, if we're talking like CPG, you can get way more flexible with this. So let's say that

instead this is just creatine powder and the website it's called creatine powder. Well, this is where we'd want to start going and doing keyword research and finding out what key term variations of creatine powder is working. And we might find that dissolvable creatine powder is a very big search key term which performs well. And so we go dissolvable creatine powder. Then we look for another key term that's doing well and we might find that it's clear and then flavored and then portable. And then at the very end we can put our brand name. And so

what we are doing is chunking in as many relevant key terms as possible into the title so that we're maximizing the surface area in which we replace on. A little additional hack here is that if this creatine powder has multiple variants, if you have let's say a 50 g version, then you have a 100 g version and then you have a 500 g version. Each of these depending on the how the feed is set up, each of these will have its own separate shopping listing. As a product of that, you can have different titles, different

photos, different descriptions for all three of these variants despite it ultimately going to the same landing page with the same product. And so you can have one of these focused on a pool of keywords. Then you can have the other one focused on a different pool of keywords and then this one focused on a different pool of keywords. So you're effectively finding different audiences and maximizing your Surface area across Google by leveraging all the different variants and spinning off the titles accordingly. On images, there's two things you want to fix. You want to make sure

that the aspect ratio actually matches the platform. There's a really common issue in fashion, which is that the images by default will be a 9 by6. And so when they pull through into the feed, they'll get autocropped. And often the autocropping doesn't actually look very good. It isn't correct. And so instead, you want to make sure that you're overriding the feed with a supplementary feed that has images in a 4x5, which is now the default image ratio that Google wants to receive. The other important consideration of images is that the idea is that the image

takes up about 70% of the actual real estate in a shopping listing. So, it is arguably one of the most important variables for being able to generate the click. Other than obvious price sensitivity on the price down the bottom and then the title, which might sway people, the image is really ultimately what's going to generate whether someone clicks or not. Now, do we necessarily want to just maximize CTR on the listing? Not necessarily. If we're maximizing CTR on the wrong people, you can think of the image the same as a hook on Facebook. The idea

of a hook isn't to hook everyone. It's to hook the right person. It's the same thing as what Eugene Schwarz says with headlines. The idea of a headline or the start of an ad is not to get everyone to read, is to get the right person to read. And the same thing with the image. So, we want to make the image as clickbaity as possible, but to the right audience. Now, within the restrictions of brand in most businesses, what that actually ends up looking like is just making the image look different from the rest of

the listings that are appearing next to it. So, how do we build contrast in the imagery that we're flowing through into the feed versus what the competitors are doing? So, if the competitors, for example, just have white backgrounds and products getting placed on a photo shoot, can we have a photo shoot? Sure, but have the background be a different color so that there's contrast built within how we actually place across Google Shopping. Besides the image and the title, there's the description and all the other attributes that need to get cleaned up within the feed. That

should honestly already be in place. If it's not, do it. It's a onetime setup and then you're good. Another thing on the feed is you do just want to make sure you have multiple feeds for each different currency. A big issue that I see in a lot of audits is that all of the different regions will just be running through one feed with one currency. Now, the issue there is that if you're placing in other currencies, it won't place in the localized currency. it would do a live conversion and show you the live conversion rate

on the shopping listing and say this has been converted from AUD or this is being converted from USD plus tax. That immediately kills click-through rates and it kills conversion rates because people know that they're buying from an international store when they might not actually be. You just haven't set your feeds up correctly. So, you finished the video, you want to go away, you want to start applying the stuff that you've learned. What should the 90-day roll out of these changes actually look like? Number one, going back to the start of the video, you want to

make sure your tracking is set up correctly. Make sure you don't have J4 events. Make sure you're triggering back events with enhanced conversions on auto taggings on. Everything's up to scratch. Number two is you want to start building out a structure map. So how many campaigns do we think we need? We always the goal is one and then what is the commercial reason for each additional campaign? Are We splitting brand versus nonbranded and do we have naming consistency across this structure that we're going to implement? Number three is you should do a Pmax specific check.

You should check that your asset groups are split by product type and that the listing groups are split out. You want to make sure your brand exclusions on. You want to make sure that audience signals and stuff are present, but it's not absolutely urgent. And you want to make sure that your target rorowaz isn't absurdly high, which is just causing it to be a retargeting campaign. You want to make sure that your bidding strategy is the correct one based on everything that I ran through in bidding strategies. So, we ran through whether you should be

expansive or restrictive and what the difference is between maxing for conversions and maxing for value. You want to go and audit and make sure there's no wasted spend across the account. Is there a campaign in there that's just spending like $400 a month and it doesn't even need to be there and you could just roll the spend up? We'll kill it. Is there a bunch of keywords that are spending on search terms that actually aren't profitable? Well, then negative keyword them. Do you have display enabled on the search campaigns? Well, then turn it off. Then

number six, you want to do a health check on all the ad assets that have been set up in the account. Do we have site links? Do we have call ads? Do we have promos that aren't stale or outdated? Are we maximizing assets within every single campaign? Because this is ultimately going to increase the real estate. So, how much space your ads take up on the landing page, which is always going to be an easy quick win to improve performance. And then lastly, we need to do a feed check. So, you want to check, are

your titles optimized? Are your images optimized? Can we do anything more here? And if we can, do this in a slow roll out. Don't suddenly roll all the titles over tomorrow. Uh, if you do that, it resets learning phases on every campaign. You'll be in a bad position. So, you want to slowly roll titles over over time. See how the campaign's reacting. See also in real time if those listings are improving. So, look at clickthrough rates, look at CPCs, look at returns, see if the changes are actually making a material impact. Because if they're not,

then we might want to test some different titles with some different keywords to make sure that we can really maximize the account. So, that's everything you need to know on Google Ads for 2026. If you're a performance marketer, email us at hiringbluensedigital.com.au to apply. [snorts] If you're a brand doing over $5 million a month, click the link in the description to book a free audit. The single most common mistake on an underperforming Google ad account is that you either number one make a change that you shouldn't have and things actually get worse, number two you

don't make a change when you actually should have or number three you make a bunch of changes but the dip actually had nothing to do with the platform in the first place. The job of this 1hour training is to teach you the order in which you investigate what change you should actually make. We'll start at the very top of the business and then we'll drill our way down each layer at a time. Ask why at every layer and keep asking why until we reach the root cause. Then you can be confident in the changes that

you're actually making on Google that they're making an impact and that you needed to make them in the first place. What this root cause analysis will actually look like throughout the span of this video is it will start at the business level. We'll look at business level KPIs, then move down to channel checks, then go to campaign types, then go into the specific campaign that we think is causing the issue. Drill into a metric, drill into a submetric, find the Cause of that submetric changing, and then ultimately identify the root cause so that we can

make the appropriate change to fix the business level KPI. Now, there's three rules before you go and touch anything. Number one is check change history. Don't ever go and do an audit or start making changes after you've already just made a bunch of changes. Also, if you're managing the account with multiple people or even if the client might have gone in and tweaked something, you need to be across it. So, make sure to always check change history before you even start this process in the first place. Number two is you want to extend time horizons.

And so, what a lot of people will do is they'll look at two short time horizons to be able to see any kind of trend. So, they end up working off just small sample size bias and they make decisions that they shouldn't. So don't look at a 1-day period of performance and then go in and go through this whole diagnostic process. Make sure that you're zooming out as much as possible contextual to the conversion volume of the account. And then number three is check conversion latency. So this is an issue that just exists in Google,

which is that when a conversion occurs, the conversion by default doesn't actually get attributed to the day in which they purchase, but instead it gets attributed to the day in which they click. And so if I just give you a quick timeline in case you're unfamiliar with this, let's say over here on the first of the month I click on an ad. Then on the 3rd of the month I click on another ad on Google and then by the 7th I actually end up buying. This is the moment of purchase. Well, what happens in most

platforms is the purchase will then get attributed into the platform on the 7th. And when you open up and you look at the seventh, you'll see oh purchase happened here. On Google instead, the purchase actually gets attributed back here. And so what then occurs is that when you're looking at the last 7 days or the last 3 to 4 Days of data is rorowaz always looks terrible. Conversion volume always looks terrible because we haven't given all of these clicks enough time to purchase in the future and then get attributed backwards. And so if you're beginner

level at Google or you're just opening up Google for the first time, you'll pretty much always see the last 30 days. If you look at rorowaz, if you look at conversion value, it'll be relatively stable and then the last few days it just drops off. And if you don't know that this is what's occurring, you'll go into the platform and go, "The last three days are a disaster. The last three days are always a disaster in every single account that occurs." Now, there is a way to get around this. And what you do is you

pull out a custom column called conversions by conversion time. So, the core keyword here to always be looking for is conversion time. Whenever you're looking at conversion time, it is the day in which they actually purchase, not attributing backwards to the click. So when you use this, you will get a better read on the last 3 to four days of real-time performance and therefore be able to make better decisions and not get looped into a diagnostic process that actually is to do with conversion latency rather than anything within the business actually getting worse. So starting

off at level one on business signal, the first thing we need to understand is Google ads actually broken and the first way that we check this is we do a cross channel check. Now, a really extreme example of this would be, let's say, 90% of your ad spend is on Meta and only 10% is on Google. Well, when revenue drops, when there's a dip in efficiency, when any top level business KPI decreases, it's probably to do with Meta and not to do with Google. Just considering that Google is only driving 10% of ad spend, probably

half of this is just branded. And so, maybe there's 5% of new customer acquisition coming through Google as a platform. If revenue dips, probably not to do with Google. So don't go and troubleshoot the platform because of it. So we need to have more of a omni channel understanding of the media mix to be able to go, okay, what's happening on the other platforms? Could something on the other platforms be causing the dip? Now let's go into a more realistic scenario where maybe you're 65 Meta and then 35 Google. Well, now if there's a revenue

dip, it's probably to do with Meta, but we can't guarantee it. Could also be to do with Google. And so we need to do our due diligence. Yes, we're going to go and troubleshoot Meta and that's a whole another process in itself. But we also need to do some troubleshooting on Google and be like, okay, does a revenue dip have anything to do with this portion of the media span. But before we even do that, we need to start asking ourselves questions on this side first, which is did anything materially change on meta, within the

business, on Pinterest, on Tik Tok. Did anything happen outside of Google that could have caused this? That's obvious. Did we come off the back of a sale? Did we do a complete creative refresh on Meta and then right after that performance fell off? Did we change the offer on Meta? Did we do something within the business that would have primarily driven this decrease in performance and broken something? Once we have answered that question, and this shouldn't be a two-c should be an investigation in itself. If your agency side asks the client, did anything happen within

the business that we're not across? We then also need to do our own due diligence. Go into Meta, go into Tik Tok, look at change history, see what's happened. Check that the website isn't broken. Ultimately understand of the revenue equation which metric decreased. So revenue equals conversion rate time average order value time Sessions. So of these three metrics, which one changed? Did average order value decrease? Did conversion rate decrease? Or did sessions decrease? Now if conversion rate decreased, we want to have a look at the website. If sessions decreased, it's probably to do with where

the traffic is coming from and one of the platforms has decreased materially in click volume. And so we need to figure out, okay, where's that click volume falling off from? If it's an average order value change, this might be a change in the product portfolio prioritization, which might be directly in Google Ads. Okay, some products might be getting pushed more than others all of a sudden. Or this might be to do with a website change in terms of a structured upsell or cross-ell that's changed materially. So we want to do all this investigation on these

three numbers, figure out which one changed, what could have changed it, does it have anything to do with the other platforms before we even get to Google. Now the prioritization of platform checking is just based on spend. So let's say that in this example actually 65% of spend was on Google and 35 was on meta. Well then we would obviously do a business level diagnostic which of the three metrics in the revenue equation have gone down and then we would check Google first. Okay Google would be the first one that we do and then we

go through this whole process. But for most people at least 80 to 90% of the businesses that we work with and that we order and that we interact with on a day-to-day basis majority of their spend skews to meta. Therefore you should do the meta analysis first and then you should go to Google. Now after you do the cross channel check and you look at the revenue equation and you understand which of these metrics decreased the next step at this level is to understand is this an attribution metric that is decreased or is it a

true business KPI. So is it revenue? Is it profit contribution? Is it me or a any efficiency level number that's indexed on the actual revenue of the business? Or are we talking about some kind of attributed number that could have some conversion time lag or where there could just be an attribution error in the setup for whatever reason tracking is broken. That's why it looks bad. Has nothing to do with the actual business. So we always want to reconcile these two numbers too in real time which is if it is a true business KPI that's

decreased which is how we operate. We would only look at this. We would rarely look at this as a means to go through a diagnostic process. And let's say revenue has decreased. Well, then the quick check to do is on all of our attribution figures, does anything correlate? What's gone down? Is meta suddenly attributing way lower and we've lost one in terms of return on ad spend. So, it's dipped from four to three. Has Google suddenly dipped off? Like what has occurred? And make sure that we're looking at conversion time, not default conversions within Google,

which will just always show that the last 3 to 4 days have been bad. And then the final level of this analysis is seasonal patterns. So understanding is the dip to do with seasonality. And this is ultimately why accurate forecasting is so critical and why it's a component of what we do at BlueSense for clients because we need to understand if revenue decreases is this in line with just the forecast and expectation of seasonality within the business. And if so, okay, cool. That is what we expected to happen. And so there's no need to go

through a 5-hour diagnostic process to try to figure out what's going on. If you don't have forecasting in place, then seasonality can just get you and you end up wasting so much time trying to diagnose what's actually going on when the reality is this is just a seasonal fluctuation that always will occur in the business. So this is level one. Ultimately, depending on how good you are, this should take you about 5 Minutes. You should be able to do the cross channel check. You should be able to prioritize other platforms first. You can look at

the revenue equation and understand which lever is causing the impact. Look to correlate attributed numbers with true business KPIs. Has it just been blatantly obvious that yeah, attribution in meta has fallen off a cliff and revenues fallen. Okay, it's probably meta and then understanding of seasonality and forecasting. So this is the first step and often you won't get past this step because on most businesses where Google is 20 to 25 to 30% of spend, Google usually isn't the reason why revenue is decreasing. In fact, Google is one of the most consistent platforms out of all

of the advertising platforms. Meta's all over the place. Okay, you do a creative refresh. You might get a winner. Suddenly, you can triple ad spend overnight. Your winner suddenly fatigues. Ad spend has to pull back. Efficiency falls off a cliff. Okay, meta is all over the place. Particularly if you don't have a consistent process in place to be able to introduce creatives within a testing structure, scale conservatively and use portfolio management when it comes to creatives, which by the way, we have a whole video on. It's 2 and 1/2 hours. It's called creative strategy in

2026. I recommend you watch it if you want a better understanding of the metaite on Google. Let's say you make a pass this step. nothing has happened anywhere else. It seems to be definitely a Google issue. Well, then that takes us into level two and three. You drill into Google Ads the same way every time, which is that you start at the account level first. You sort by campaigns by cost descending. So, you want the highest spending campaigns at the top. The campaign you're looking for is somewhere in the top five by spend. Below the

top five Spending campaigns, even if you doubled performance on them or performance fell off a cliff, it wouldn't actually materially move the account. I'll give you an example. If you have a campaign that's holding, let's say, 10% of budget, even if you got a 100% increase in performance on this campaign, it's not going to significantly impact the account really at all. It's going to have a 5% impact. And then if Google is only 40% of your media mix, we're talking about singledigit percentages. And so, if there's a large material change at a business level, it

is not to do with a campaign that's holding 10% of your Google spend, which is why you want to go straight to the top and start at the highest spending campaigns. Now, is there a reality where this could be the reason why the business dropped? For sure, but it's just not likely. And so, we want to start at the most likely reasons as to why performance has dipped and then move our way through into checking all the small things that realistically is probably just a waste of time. Hence why we want to dep prioritize them.

Here's the 5minute check process. Number one, take your date range and look at the last 90 to 180 days and switch to a weekly view. Then you can just stay in the overview tab of Google, so top left, and you can just look at all the graphs. And what you want to do is rotate through each different metric here and understand how it's moving. You want to look at cost. Has cost materially changed over the course of the last 90 days. Conversions, has conversions changed, conversion value, cost per conversion, return on ad spend, which is

conversion value divided by cost, and then click-through rate. Have any of these metrics materially changed when you're looking at them in a graph view over the course of the last 90 to 180 days? What has gone up? What has gone down? What has gone sideways? From there, you want to go away from the overview tab and you want to go into the campaign tab on the left. And now we want to drill down at an individual campaign level. And so we're looking at each individual campaign. Once again, the top five campaigns, the top five biggest

spenders, start at the top, work your way down. Which campaign has contributed the most to these metrics changing. So let's say conversions has slowly pulled off and it happened on a specific date. So you want to identify when is the inflection point? Is it the 3rd of March? And then from the 3rd of March onwards, did it start trailing off? And what you'll find if this is a Google Ads issue is that maybe there is one or two campaigns where conversions or rorowaz or click-through rate started declining after a particular date or there was a

material impact. From there we drill one step deeper which is we then go down to the ad group level. If there are multiple ad groups or in Pmax asset groups we do the exact same exercise. We look at all of these metrics. We look over the time period and we go which asset group or ad group has contributed to this decline in the metric. Is it all of them? Is it just a specific one? Cool. Now we know exactly what in the account has caused a material change in a topline KPI where we can now

start to move through and identify what metric has caused it, what submetric and then what is the root cause. Now the real two keys that we want to know off the back of this is number one, what campaign type has fallen off? Is this Pmax campaigns? Is this shopping campaigns? Is this search campaigns? Is this display? Is this YouTube? etc. and then off the back of that which specific campaigns and which specific ad groups. Now the reason why the campaign type matters is because the Failure modes of each campaign type is different. So a Pax

campaign will decline for different reasons to a shopping campaign. A shopping campaign will decline for different reasons than a search campaign. And so it's important for us to understand the campaign type and then we can go into troubleshooting it. So what are the specific failure modes of a PMAX campaign? Number one is warm and cold drift. So because performance mass campaigns can retarget people and it can place across all the different channels. What can happen is Pass campaigns can start retargeting people more or start going into cold targeting more dynamically at its own will. As

an example, the campaign might be labeled a cold campaign, but it now starts serving mostly to existing customers. And so you want to go and check the search term report. Now brand should be excluded anyway, but it's good to just double check. And then number two is you want to go and check the audience report. If brand terms have started to get introduced into the campaign or if the audience report shows a skew towards existing customers then the brand exclusion is broken or the target rowaz has been pushed too high on the account on this

campaign sorry which is causing it to rep prioritize warm audiences. Number two is feed disapprovals. In e-commerce specifically 80 to 90% of the performance in Pmax is going to come from shopping and shopping performance is going to come from feed quality. And so if something has gone down in the feed, if a product has gone down, if there's some kind of disapproval that's occurred, that's obviously going to be a main contributor to why the campaign's performance started to decrease. So as a product of that, you want to go and look at the feed. So you

want to check products just within the campaign, see if any top spending products have gone out of stock or have been turned off or being disapproved. And then number two is go into the GMC as well and just double Check everything there. Number three is high rorowaz narrowing the targeting. This is something that we've gone through in the other Google video that we put out. But as you increase target rorowaz on a campaign, it doesn't magically mean that you just suddenly get better efficiency on a campaign. That's not really how it works. What's actually happening

is that Google is narrowing the targeting to a smaller subset of buyers that are more likely to convert at a higher efficiency. So, it's only going to enter auctions where it knows it's going to win and that person is a high likelihood to purchase. When it does that, what typically happens is that you're just narrowing in on a warmer audience. And so if you start increasing and increase and increasing target rorowaz, it's just going to narrow the pool of audience that you're targeting. Spend will likely pull back and you'll probably get worse new customer acquisition

in the campaign. You can actually double check this with third party attribution tools these days. You can also just see it natively within Google ads too. But if you look at any NC rorowaz numbers on like a triple whale or any tool that you use, uh NC rorowaz numbers on a pmax campaign with a high target rorowaz will typically be really bad. You could have the exact same campaign and have a low target rorowaz and NC rorowaz will be better. And so the actual target rorowaz is not indicative of the performance of the campaign on

cold audiences. It is actually indicative of how hard the campaign will go on just retargeting warm people. And then number four here is product bloat. What this means is that and this isn't going to be applicable to most people but if you're rapidly increasing the skew count on the website and all of these SKs and new products are flooding into this campaign, you can just end up with so many products in the campaign that it impacts learning. A subset point of this Is that if you have products going in and out of stock all of

the time, that also resets learnings of the campaign, particularly if it's a high spending product. Let's say a product is holding 15% of total spend in the PMAX campaign and this product went out of stock for 4 days and then come came back in stock. This is not a good position to be in because the learning phase of this individual product resets, but it also impacts the overall campaign as well. And so we actually have in one of our onboarding videos for clients a disclaimer around this exact point which is that if you have products

that go in and out of stock all the time, please let us know because it might materially change the way that we decide to structure the account because we don't want one product going in and out of stock impacting the performance of all the other products that sit in the same campaign. If it is a search campaign that has failed, number one, you want to look at search terms. So what search terms are we spending on and has this materially changed over the course of time from the inflection point? So there was a point in

time in which the campaign stopped performing. What happened before and after in the search term portfolio of key terms that are getting most spend and has anything materially changed? So you go into the search term report of the search campaign. You look at the time period beforehand. You sort by spend and you go okay what are our top spending search terms here and what was efficiency and now we look at after that time period and we go what does it look like now? Has it materially changed? If it hasn't materially changed if all the search

terms look similar performance looks similar across them then you can move on. But often what can be the case particularly if you're using broad match and smart bidding strategies is it might be a wild change in search term uh spend allocation which has actually caused the impact in performance. Number two you want to look at auction insights. It might just be the case that some competitors came in and launched search campaigns and have started to drown you Out of the auction which has increased CPCs or has decreased your ability to spend because you're no longer

entering auctions correctly. Um you're sorry you're no longer entering as many auctions. Number three is target rorowaz. Exact same thing as with pmass campaigns. If you just started to squeeze this up, it would have narrowed the audience targeting. Number four, you want to make sure display is turned off. Or else you could have just had spend getting allocated into display which was causing a performance drop off. In regards to shopping, it's all the same stuff as Pmax except for the warm versus cold. So you've got Troz, you've got GMC. This is likely the biggest and

most important thing to check, right? Has the products gone out of stock? Did we change titles or descriptions recently? Is there any kind of errors or warnings within the GMC feed? Do we have now too many products flooding into the campaign? One that's probably not causing anything, but it's worth checking is is there a lot of spend getting distributed into search partners and is this performing poorly? That then takes us into level four and level five. The important concept to understand at this stage when we drill into a specific campaign and we start looking at

one metric is that every campaign level KPI that you care about like revenue like rorowaz like even conversion rate is the product of two or three underlying metrics. And this is why understanding the formulas that constitute every metric in e-commerce becomes really helpful for being able to troubleshoot and do bottleneck analysis. If you understand for example that average order value isn't just average order value but it is a blend of new customer Average order value and returning customer average order value and then you understand that each of these sits on distribution curves and then you

understand that the actual mechanics that impact each individual order here is units per transaction and average unit retail then all of a sudden you can go through a troubleshooting process on average order value that is much more in-depth comprehensive and aligned to the actual root cause than anyone else because some people will try to troublesoot average order value by just looking at oh what offers change but if you understand that no actually we need to drill into NC because that's the one that dipped then we need to look at the distribution curve and how that

changed and then this part of the distribution curve changed what actually caused the change was it units per transaction or average unit retail was up what mechanically caused the dip in and then we can actually troubleshoot this which is the root cause this isn't the root cause and so the deeper you can go in your metric understanding ultimately the better you will be at being able to do root cause analysis and actually troubleshoot the business and So in Google Ads, if revenue is down, attributed revenue, it's because of cost per click conversion rate and average

order value because that is ultimately the revenue equation, but we're taking clicks and we're going into CPC in the platform. If conversions are down, it's because either clicks drops, conversion rate dropped, or tracking broke in some way, and that's why the attributed conversions aren't there anymore. The job at this layer of troubleshooting is to figure out which of the underlying metrics moved because each one points to a different cause. There's really two formulas that you want to keep top of mind. Now, I could write infinite here and every metric derives into submetrics, but these are

the two that you want to keep top of mind because this will get you 90% of the way most of The time, which is CPA equals CPC divided by conversion rate. So, these are the two submetrics of cost per acquisition. And then clicks equals impressions times by click-through rate. And so if click volume goes down, it's a product of either impressions compressing, which means we're entering less auctions and showing uh less of a degree, or click-through rate is down, which means that other people's listings on shopping on search, whatever, are suddenly more convincing than ours,

and so we're losing out on the actual clicks in the auctions that we're entering. So these are the first two that I would actually start with in terms of troubleshooting, which is, is our rorowaz and efficiency down? Is that the issue? Are we seeing a drop off in performance, but volume is relatively the same in terms of click volume? Okay, cool. Then we need to drill into CPCs and conversion rates. Now, what are the common causes of CPCs being down? Number one is that your quality score could have materially changed. The way that you check

this is you open up keywords, you add quality score as a column, you open up keywords, and you add the following columns. You add expected CTR, you add ad relevance, and you add landing page experience. If any of these three have materially changed post intervention, so post the point in which performance dropped off, then that is the thing that you need to troubleshoot. Ad relevance is often a really easy problem to solve. You just make need to make the ad more relevant to whatever search terms are actually being placed on. Expected CTR is a product

of actual current CTR and then it's just extrapolating it into the future. And so this just means that your ads need to be better and you need to improve them. uh landing page experience is actually the Hardest one to materially change because it actually requires dev work and the reync cycle on this. So how often Google will scrape the website and then actually change this score is kind of pretty unknown. Uh I've been in positions myself where landing page experience has been the bottleneck and what actually caused a dip in performance and it's taken weeks

if not months to actually be able to turn it around. We fixed everything on the website but Google just wasn't rescanning uh the website and rescoring us. Now number two is a new competitor could have entered the auction and this is driving up CPCs. And so the way to check this is really easy. You just go into auction insights and you can look at auction insights prior to uh the point in which performance dipped and then auction insights after the point in which performance dipped and then you can see has someone new gone and entered

the auction. And then number three is you might actually just be placing on premium auctions. What this means is that you might have made a change to target rorowaz or any of your bidding strategies and as a product of that you are now entering into more expensive auctions. So it's becoming more expensive to get clicks. Not necessarily a bad thing because they might be higher quality clicks. And so this is actually an important thing to understand about CPCs in Google ads is that it's relatively a vanity metric. In fact, it's kind of a vanity metric

as well in meta and Tik Tok. And the reason being is that it's not directly associated with performance except on its fringes. So if CPCs are super high, yeah, it's an issue. If CPCs are super low, yeah, it's an issue. But CPCs can go up and as long as conversion rates go up as well, we're all good because CPA stays the same. The issue becomes when CPCs go up and Conversion rates stay the same, then it's like, oh, okay, well, it might be a quality score or competition or a premium auction issue. and more specifically,

we're entering premium auctions, but it's not paying off. We're not getting higher quality clicks and higher quality users to the website, and so it isn't worth it. Now, on conversion rate being down, the first thing that you want to check is just tracking. Make sure has tracking stopped working. Are you looking at the last short time period, but you're not looking at conversion by time? Those are going to make conversion rate look bad, but it actually has nothing to do with anything in the platform. Number two, has the landing page materially changed? Now, number one,

have you actually changed to a different URL? Number two, have you made changes materially to the URL? Has the price changed? Is there some kind of redirect in place that's now pushing to a different website? Has there been upsells or crossells added? Has the merchandising on the website changed? Ultimately, if conversion rate is decreasing, this is probably the number one reason as to why. Number three, has the audience changed? You can have the same landing page, same website, but if you have higher quality users coming from more premium auctions, your conversion rate will be better.

vice versa. It can be worse if you have worse quality audiences. So, you want to look into search term reports there. Have the search terms that we're driving people through changed? Are they coming from different keywords that might not be converting as well? Because maybe there's not congruency there with the landing page. Have we materially changed the bidding strategy that might be pushing towards and optimizing towards a different audience? What's going on here? Number four is unfortunately the learning phase is a thing on Google Ads. It's actually a very big thing and a very annoying

part of the platform. And so, has the learning phase reset in some capacity? Have we made some large structural change to the account that's Pushed it back into learning phase which will always decrease conversion rates because you're going to be entering into the wrong auctions and be driving poor quality traffic to the website. Number five, and this is off really the back of the landing page, but has the offer changed? Has a promo ended? Has a discount being removed? Has a free shipping threshold changed on the website? Because this is also going to impact conversion

rates. Now, let's say it's not a CPA issue. And so, our efficiency is the same. It hasn't changed, but instead volume has changed. We're now either not spending as much or we're not getting as much revenue. Well, that is usually a function of clicks falling off. We're driving less volume out of Google Ads to the website and as a product of that, we're doing less volume. So, the subset of clicks, the submetrics is impressions and click-through rate. On impressions, there's two types of reasons as to why you lose impressions or you lose impression share. Number

one is impression share lost to budget. You just don't have the budget and so you can't get more impressions cuz you're maxed out. there's no more spend to be able to get you more impressions. Number two is impression share loss to the bid and so your bid just isn't high enough to enter enough auctions to actually spend the money. You want to identify which one it is and the way to identify it is pretty easy. You're not losing impression share to budget. Sorry, you are losing impression share to budget if you're hitting the budget. So

if you have a $100 a day budget and you're hitting it, well the reason why you're not getting more impressions is likely because you just need to increase spend. Now you might be thinking our spend has always been the same. Why has our impressions changed? The impressions might have changed because now you have higher quality users. So CTR has gone up which is offsetting click volume. Now if click volume is down and impressions are down, I can almost guarantee it has nothing to do with your budget because Your impressions and click volume is now getting

throttled likely due to the bid. Now the bid is a product of all three of these things. It's a product of quality score, new competitors, and how aggressively you're entering auctions. You want to go back and troubleshoot all of these to be able to fix impressions I lost a bit. On CTR decreasing, it's usually one of two things. It's ad fatigue, which is pretty rare on Google to be completely honest. And then it's an audience mismatch. Now, why would the audience be correct historically and now be bad now? Like why did we used to have

high CTRs but now CTRs are declining? Well, it is because of over here the audience changes. So, you can see all of these are somewhat related. Anytime you troubleshoot a submetric, it is going to be related to a similar process that you do elsewhere. If conversion rates decrease, it's likely due to due to an audience change. If CTRs decrease, it's likely due to an audience change because you're taking the ad that used to resonate with people very well on specific search terms with specific people and now it's not and CTR is lower. And so that

is probably a product of the audience that you've been targeting materially changing. It could also be fatigue on the ad. Now, this is super unlikely in something like shopping ads. In search ads, it can potentially be the case. Um, you could just have incorrect seasonal language on the ad. So, you could be talking about some kind of summer sale, but it's no longer a summer sale. So, you want to be just looking at the ad copy. Is there something here that lacks congruency to this moment in time? Um, another reason why CTRs will decrease quite

heavily on Google Shopping is due to competitor pricing. And so, you want to look at competitive pricing on the products. And that will normally tell you one for one CTRs have decreased because one of your competitors has gone on to a 20% off sale. And so, now anytime someone is searching for Nike shoes, they're not going to click on you. they're going to click on the competitor cuz they sell the same product for way cheaper than you. In fact, we actually found this with uh one of our clients that we work with where they are

a large wholesaler of sunglasses and every time performance decreases at a product level, it's a product of their biggest competitor taking that product to sale on Google Shopping. And so we set up a real-time script to be able to understand price changes on the competitor website so that in real time as a competitor would drop prices, we would just pull the product or depp prioritize the product within the campaign structure. And the reason being is we could continue running it and continue forcing spend to it, but it would never be profitable because row would just

drop off a cliff the second the competitor goes in and just wipes price by 20 to 30%. So rather than playing that game and meeting them on discounts, which would have eroded gross margin to almost nothing, instead we're just not playing the game at all and we just pull the products out of the auction the second the competitor starts discounting to no margin. Also, as a side note before we move on, you can actually check this as a metric in Google. It's called search impression share lost bracket rank. So I recommend pulling this out and

adding it to your columns. Now on to level eight, which is the root cause. The default way to think through a next step here on Google specifically is that you always want to avoid a rebuild or a restructure because anytime you rebuild it triggers a learning reset which is going to just put you in an even worse position. And this is where you can get yourself a really risky position on Google which is that performance declines. So you make a big change which causes performance to decline even more. So you make a big change which

causes performance to decline even more and you get into this Feedback loop until you're in a terrible position. And so you want to be very careful and the first thing that you want to do is just apply a fix. So let's say we drilled all the way down to the submetric and we found out that conversion rate on the landing page was the issue that caused all of these upstream issues. And so as a product of that we apply a fix to the landing page. We make a material change. Then the key here is that

we need to be confident in allowing enough time to then see if the fix worked and then potentially go into another fix. We also need to have the understanding of whether we can group multiple fixes at once and that it won't confound the outcome in being able to draw what actually worked. And so this is where you have to be incredibly strategical. And this is probably the most strategical part of this entire process because we might find that conversion rates are down on the landing. There's a million things that we can do to improve conversion

rates, right? We can change the offer. We can improve the landing page design. We can put more continuity into the ad traffic that's actually driving here. There's all this stuff that we can do. But if we do it all and it works, we don't actually know what worked, what materially changed if we did nine things. And so there actually is value in stripping back and only doing a few things so that we know what actually made the impact. Or else we're just throwing the kitchen sink at a problem and then the problem solves and we're

going nice. But then when the problem occurs again, we don't know how to solve it without throwing the kitchen sink at it again. So this is where you have to assess the severity of the situation and whether we want a strong conclusion off the back end of the fix. Sometimes we might be in a situation that is so dire and so bad that we do just need to throw the kitchen sink and we don't care what works as long as it's fixed. Sometimes it's just a little bit of a performance decrease and we actually want

to Understand what the fix is so that in future it doesn't happen again. And so you need to think through the prioritization of the fixes that you're going to go and apply. Now, I can't really give you the fixes in this video because there is like 20 different submetrics that could be impacted. And off the back of those 20 different submetrics, there's probably a hundred different potential fixes for each individual one contextual to the particular business. Which is why your ability to do root cause analysis and then actually apply fixes um becomes probably the most

important skill in performance marketing and media buying because there is such a large decision tree in all of the different things that you can do that really good media buyers and really good performance marketers are good at being able to identify the ones that will actually fix the problem. The main thing that I would recommend just thinking through when you're thinking through the fix here to be able to solve the submetric is how much of an impact is this change truly going to make? And you need to be unbelievably honest with your answer to that

question. Is changing the headline on the website really going to change conversion rates and fix this entire business? Probably not, unless they had a different headline previously and there was a headline tweak which actually caused a decrease in performance. Okay? Okay. And so we need to think about the severity of the expected impact of the fix that we're applying. And if it's not enough to fix the decline in performance, then we need to continue to rethink the fix until we believe it will be strong enough to get us out of the problem area that we're

in. So if there's three things to take away from this root cause analysis process on Google, it's number one, don't change anything before you understand it. You want to check change history first. You want to make sure That you're checking latency in the way the conversions are getting attributed into the account so you're accommodating for conversion time. You want to do a cross channel sanity check first. So you don't just want to rush in and try to fix Google when it might be something else. Because the reality is is that most accounts that appear broken

aren't actually broken at all. You're probably just misreading them or looking at the data wrong. Number two is you want to drill from the top down. You never want to start at the bottom and try to work your way up. And this is really the core reason why most people are not good at root cause analysis is people go straight to the bottom and try to solve hair either as a time-saving exercise. I honestly don't know why people do this but you need to start at the top. What is materially changed in the business? Because

you might have a client come to you and the client goes the rorowaz on this campaign is down. It's like okay rather than going there to start zoom out go to the top. Has anything at a business level changed materially at all? No. Everything's actually up. Okay. If everything's up, does it matter that this particular campaign has a 10% decline in rorowaz in the Google account and it's the fifth biggest spender? Probably not. It probably doesn't matter at all. Now, should we still troubleshoot why it was down? Sure. But let's start at the top first

to understand the severity. Then let's go to a channel level and go, well, has anything materially happened on the other channels that could have impacted this campaign? Because this might be a brand search campaign which is just completely dependent on metas-pan. Then let's go into the campaign type. Then let's go into the specific campaign. And then let's finally get to the submetric that the client pointed out and see now that we have all of this larger context Within the business whether it matters at all. And we might find out that it does actually matter and

it does connect to the top and therefore we need to go into a root cause analysis and make a material change. But a lot of the time a submetric will get called out to you and like oh why CPC's down in this particular ad. It's like well hold a second zoom out go back to the top. Is this impacting anything at all? Does this actually matter? Is this worth me spending an hour of my time on? No. All right. Well, then let's not spend an hour of time troubleshooting this one particular item. My CPCs are

down when it's absolutely meaningless to the growth of the business. Or we go to the top, we work our way through for 5 10 minutes and we find actually yeah, this does matter. This is a great call out. All right, let's try to figure out why did CPCs decline on this particular ad and how do we fix it? So, always go back to the top. Don't let yourself get drawn straight into the bottom. And then number three takeaway is that once you go through this whole process and you get to the bottom, you want to

make sure that you're defaulting to a fix rather than a rebuild. Rebuilding ends up costing you the learning phase and will likely just make things worse. Rebuild is the last thing you really want to do. You always want to start at what are the fixes that we can apply? What are the highest impact fixes? Will it make the impact that's required to change us back to baseline? And only if the fixes don't work or if the fixes won't make a material enough difference to be able to bridge the gap to the historical performance, only then

do we go to a rebuild. One of the most dangerous things that performance marketers do is that they tweak all the time. They see a number drop and they reach for a setting, but most of the time it just makes the problem worse because the setting was not the cause and the change further moves things in the wrong direction. The skill that ultimately Separates a seriously good performance marketer from someone who's just tweaking against metrics all day is the order of investigation. They start broad. They drill narrower. They ask at every single level why. And

then they only act when they can actually finish the sentence. This campaign is underperforming because this metric drop driven by this submetric which was caused by this trigger which itself was the root cause. If you can't put that sentence together after this process, you do not touch the account. You need to be able to say which metric dropped, what was the submetric, why did that submetric change, and therefore what is the root cause. If you're a performance marketer and you found this video helpful, please reach out to us at hiring at bluedigital.com.au. We are always

hiring for more talented performance marketers. And if you're a brand doing over $5 million a year in revenue, click the link in the description. It will take you to a short two to three minute video which runs you through how our audit process works.