The REAL Reason Your Facebook Ads Won’t Scale

You don't have a creative problem. You don't have a media buying problem. You don't even have a messaging problem. You have a personas problem. Yes, personas are the very foundation of Facebook advertising in 2026. And Metra have come out and said this. If you're experiencing incremental reach plateaus and you keep throwing more and more creative volume at the problem, you're just going to be hitting the same audience over and over again. So, if this is the case, why do most brands still not understand their personas? Because a lot of people when you ask them what

their persona is will say it's a 45 to 55year-old mom who has three kids in Denver, Colorado. Like that is not a persona. A persona is not a customer segment. It's not a brand archetype. A persona is about the psychoraphics, not the demographics. To understand a persona is to understand how someone makes decisions, what they are afraid of, what their core values are, and what is going to drive them to pull out their credit card. So, what I'm going to be walking you through today is how to create a persona analysis document, how to go

deep on your core personas and how that should impact the ads that you create. All right. So, I'm going to split this video into two separate parts. We're going to start by identifying our core personas and then we are going to examine the ad account to uh see if we can find any underserved personas based on our voice of customer. Here is the document that we're going to be walking through today. I have put this document in the description below if you want to go and apply this process to your brand. I'm not saying this

is the way to make persona just how we have done it and it is a deeper process that I than I have seen shared online elsewhere. Um so let's get straight into it. Um step one persona identification. This is where we're going to list all of the different personas that we uh potentially have and then we're going to identify what are our core personas. Now, I have given you prompts for this should you want to use them, take them, adapt them, or just put them straight into Claude. I use the text stack for this. I

use is Claude and Parker. I'm going to show you how I use them together cuz I don't just rely on one or the other. If you're unfamiliar, Parker is the uh AI tool that we've built uh at Adrete to help you make better decisions inside your ad account. It connects to all of your different data sources and basically acts as you know your AI creator director helping you understand what ads to make. We're going to be going through that. If you don't have Parker, this is also going to be applicable to you uh because you

can you can do this inside of claw. It's easier in Parker. But as you'll see, we use both. So what I'm going to do is get Parker to look at all of the different data sources has access to and start to list all the different personas that we can run with. So, I'm going to take this prompt and I'm going to put it into the chat. Uh, today we are going to be advertising for Laura Gella, which is a makeup brand for mature skin. Um, so I am going to enter this prompt in. I'll talk

you through the prompt in a second. I'm just going to go and let the prompt run. In the meantime, let's go and grab the clawed prompt and drop that in there as well. It's just basically going to look through all the different data sources. Parker uh has more like access to the like more private data sources like uh you know ad comments has access to obviously the ad account uh competitor ads uh posture service you can put inside of there as well customer reviews obviously claude uh does better at the web scraping so this is

why I like to use both uh because they both have their merits and sometimes yas are better inside of parker sometimes they're better inside of claude but when you combine them that's when they get super strong so let's put this prompt inside of Claude and I'll be back in a moment. Okay, so we have our outputs. Let's have a look inside of Parker first. Uh I've structured these prompts in such a way where it will give me a lot of other information. It will always uh basically fact check uh that it's giving me the right

reviews, comments, etc. Uh I like to get it to give me numbers as well. So for example, where I can see that Park has analyzed 6,000 reviews, 9,000 comments. There aren't any posting surveys in here yet because we only add that the other day. um and no other information, but you easily could add that in here um if you wanted to include that in this analysis. We have a few different distinct films that Parker has analyzed. Now, it's important to know that the first step is just to list the different personas. Then, we're going to

go deep. And one thing I want you guys to get from this process is that I am not trying to give you one prompt and then just you you run it and that's your personas. I do want you to challenge Parker. I do want you to challenge Claude and kind of get them to work together alongside you to develop your personas because regardless of how long you know the prompt is or how much time that we've spent making it, this this prompt had gone through a lot of iterations, sometimes the personas that it generates or

some of them will still not be spot on as you'll see in a second. Um, but when you use a combination of them plus your prompting, you can get to what I believe to be the most accurate uh personas into your persona document. So, the personas that we've got here is the woman whose old foundation betrayed her, the woman who's quiet wearing makeup, uh the daughter who's found something that might finally might help her mom, um and the gal. So, like a couple of these, this down the bottom is more of retention um uh persona

than an acquisition one. I think these two top ones are strong. You can go into the proof quotes and again, this is where I think Parker is stronger. Parker's data sources are richer um whereas Claude just relies on web scraping unless you go and build something out where you upload uh all of your reviews uh post purchase service etc inside of claude but I have found pers to be better at this side of it when it comes to actually extracting support and evidence to back what it is saying um going to give you the buying

situation proof quotes etc. Uh and um then it's going to say what it didn't include. Now, I do think it's also important here to talk about what a persona is not because there is a a a huge amount of confusion when it comes to personas uh in the space right now. And I can tell that because of the consulting calls that I've had booked in in the last couple of months, almost all of them have involved uh persona analysis or like the idea of understanding their personas better to some extent. So, a few things that

you want to be very cautious of when it comes to personas. Number one, obviously not labeling demographics as a persona. Again, like we said earlier, a mom 45 to 55 years old who lives in Colorado, that's not a persona. Uh, another thing I want you guys to be careful of is what I call identity overlays. So, I see a lot of people say, well, and sometimes Parker and Claude will say a gifter, like the gifter is a persona. I would not consider that as a persona. Uh because a gifter can affect can technically be applied

to a lot of different personas. They can a lot of your individual personas as they are can have can be giftifters as well. People can be gifted it for that person. So I don't think that's enough uh to warrant its own persona at least not as a core persona. So you just want to watch out for that. Again, the questions that we want to answer to warrant it being a persona is what do they already believe about the problem? Uh what have they already tried and what failed? Uh what are they afraid of and what

they secretly want? And what exact words they use when they talk to themselves about this. If you can't answer those four questions directly uh about the persona. I do not believe it to be a persona. So there's a couple things you might just want to watch out for here because it can still happen inside of Parker. And I just wanted to take Parker as Parker's word as gospel or Claude's word as gospel because you have to use a combination of the two. So what I actually like to do and and actually before I do that

I probably should go into the Claude output. It was similar. Uh Claude actually gave the reluctant returner uh sorry the liquid defector which is very similar to the one the ones that Parker pulled out. Um skeptical scroller and the reluctant returner. Again I don't love like this tells you how they came to the ad. doesn't tell you about their psychographics or like what happened like what they believe about the world. So I'm not massive on this and again turn out it's not the strongest of personas uh either. So, what I actually like to do is

I will feed uh one platform the others response and say, "What do you think of this?" And I basically just try and like mash them together and work out which are the best bits from each one, which I don't know if this is the best process or not, but this is the way that I found it to be good because sometimes Claude outputs Claude's outputs are better. Sometimes Parker's better. So, what I'm going to do is I'm going to say here's the persona output they got from Claude. What are your opinions on this? I'm going

to paste that in. And while that's running, I am going to say here are the persona outputs from Parker. What are your opinions on this? We'll read the output. We'll take the best bits from both and then we move into the deep persona analysis when we are going to make a document uh detailing each one of uh these personas. Okay. So Parker and Claude have evaluated each other's work and uh yeah, some pretty interesting things here. What you'll often find is that they both find similar things uh and they just present them in different ways.

So doing this allow them to kind of riff with each other uh gets you to what I believe to be the best output. So for example here, the reluctant returner was a persona that Claude identified that Parker didn't really always said it was a small persona. the liquid defect. They both identified. Uh so those have become our two uh core personas. The daughter who found something that might help a mom like that is a little bit of a uh gifting persona. So I'm not massive on it. And the skeptical scroller, I am curious to see

how this comes out. I'm not the biggest fan of it, but I feel very good about these top two, which we will go into in a second. And Claude said the same thing, by the way. Just basically agreed that these two were the top two personas to run with. You can go back and forth with this and you can you can probe it in any way you like, but I found that just kind of getting the two to riff with each other uh results in the best uh overall personas and it really helps because they

both have their own blind spots. Claude can be better at reasoning when given the information. Parker is a lot better at surfacing evidence when it has the personas identified. So, it's more grounded in your actual voice of customer. Anyway, let's now go on to step two of this process, uh, which is the deep analysis on each one. So, what I'm going to do for this is I'm going to start a new document. This is exactly what I would do if I'm doing persona analysis for a new brand that comes on board with us. And I'm

going to run this persona analysis on each persona individually. So, I've built a long prompt here is a 10step process that you're going to go and run on each persona. You'll want this analysis to be as close to the voice of customer as possible. You can do this inside of Claude um by all means, but I would recommend doing it inside of Parker. So, let's take this prompt and we are going to update the persona name. Let's do it for uh the reluctant returner. And I believe that is the only thing that we need to

update throughout the prompt. I'm going to turn this off. Now, I want to reiterate that you use this process as a baseline rather than as your actual persona analysis. I mean, you can use it as your persona analysis, but I would love for this to be like the skeleton for you to then go and layer on top your manual research. So, I've put down here all the different data sources that we use for our research. We tend to get the things that are heavier web scraping from Claude, the things that are more uh, you know,

richer data just plugged like straight from Parker. Um, so for example, customer reviews, ad comments, support tickets, postpurchase surveys, all that stuff is inside of Parker and then you know, Trust Pilot, GC app store reviews that will come from Claude. What is difficult to get from either and where I spend a lot of time, my manual research is comments. So, Tik Tok comments, uh, Instagram post comments, Facebook groups, uh, Facebook comments, um, corer questions, which I guess you can get inside of Claude. Uh, but it's difficult for either platform to scrape comments. There's so many

golden nuggets and so much you can learn about your audience uh their pain points, their frustrations, and their desired transformations by seeing the language they use in comments. It's also pre- purchase which gives you a different uh perspective because a lot of times we just look at customer reviews or post-purchase surveys which is people who have already bought the product and in some cases people have already used the product for a long time. It's also equally as important to understand people who are uh pre-purchasing um which in many cases in the comment sections they have

done. So, I'll give you an example of that in a moment. Um, but for the time being, I believe our prompt has been returned. So, I'm going to walk you through each step of this process. You can see the full uh descriptions uh in the prompt. If you want to look inside of here, it explains what each of them are and uh why they uh need to be inside this analysis. Now, I have really tried to root this process in direct response. So, it goes very deep. You might think unnecessarily deep, but if you go

through this properly, you do this research process properly, you will have a better understanding of your audience and your personas than any of your competitors. I I guarantee it. So, let's go through it for this persona. Okay, the reluctant antenna. First of all, the Gary Halbert paragraph, which is the overview of what the persona is um and about them, just so you can understand very quickly. This is what they are uh based on quotes. So this is pulled from uh your voice of customers. So your reviews, your post purchase surveys. And again, this is what

Parker is really really good at. Better than Claude um objectively uh at servicing the the actual language these customers are using because the copy is already written in our prospect's head. We are not trying to convince them of anything new. We are just trying to identify what they fear, what they want, what they desire uh and channel it towards our product. And I think that's what Kapaka does a really good job at doing. the five Fs, fears, frustrations, failures, fantasies, and failed solutions. It goes into all of those. You guys are probably familiar what they

are, so I don't need to spend too much time inside of this, emotion, and trigger moment. This is really important to understand what people were feeling as they actually pull out their credit card. Again, it goes back to what we were saying earlier. We're not labeling our personas by their demographics. We're labeling them by what they believe about the world and like what actually would convince them to get out their credit card and pay and understand the emotions that they're feeling at that time. um is very important to be able to do that uh in

your ads. Next, we have the awareness and market sophistication levels which are both very important. This looks at all of the voice of customer and basically infers at what awareness level your audience are and that is really important for you to meet them where they are at in the process rather assuming that your audience are like aware of you or aware of your solution or aware of your problem. You can just see where they are already at and then tailor your creative to that. You want to make sure that you have creative for all the

different awareness levels because you're not going to have all of your customers in problem aware or all of them in solution aware. They're going to be distributed in different awareness levels. That's why it's important for you to understand what they look like so you can craft your uh creative per persona based on the awareness levels of the prospect. Market sophistication levels isn't something I hear talked about as often as awareness levels. Uh it's another Eugene Schwarz concept. And basically you think it like this the awareness level is where the prospect is at. The market sophistication

level is is the where the the market is at as a whole. The best example that I can think of of this is eight when they launched the smart mattress. I can't remember when it was uh I think it was like 2019 2020. Um and at the time there were barely any other competitors for this product. So the market was in stage one maybe stage two level of sophistication and that meant they could make very simple but bold claims and that would be enough to get attention. They could say for example uh the the world's

first smart mattress but now if you wanted to make that same claim in 2026 it's going to be very different. If you want to say for example the the world's best smart mattress everyone's saying that the market sophistication level has increased. It's now four or arguably even five. So you need a lot more extensive copy about why you like what's different? Is there a new mechanism? How is what you've got different to every other brand that's making this same claim? So your prospect's awareness level can stay the same yet the market becomes more sophisticated and

that would require drastically different copy than a less sophisticated market. I actually saw this on uh Google when I was doing some research in preparation for this video. So I asked Claude to turn it into an artifact. I will brush this up and put it into uh the description at some point, but it just basically talks through how to structure your ads based on the awareness level and the market sophistication level of your prospect. So, I thought that was pretty cool. I'll put that in the description below. But anyways, moving on. Let's continue rolling through

uh the Pora analysis, the buying journey. Uh we actually use multiple different uh inputs for this, so don't just rely on the AI prompt. Uh the post-purchase surveys are what this mainly based off of. As you can see here, um, this doesn't have any postparture surveys in Parker. So, uh, that's why I wouldn't rely on these specific outputs. I also use, uh, the attribution, um, inside of Triple Well, and I sometimes quiz MOI on this because it gives some interesting insights on the bio journey that people go through. So, a combination of those. This is

probably the section that I would uh work on the most outside of uh Parker um as it pertains to this specific process. Cognitive bias map to help understand the biases that the customer has in their head or the persona has in their head before buying objection stack. Uh competing solutions just going deeper on the five Fs and again some more copy phrases that appear in creative uh source from the VOC. I wasn't going to keep this in here, but it just comes out with some great lines sometimes. I just thought it's it's useful to have

in here. What I've also done inside of this props and something I would recommend you doing in Parker Claude or whatever inside of uh anything that has a a long output like this, there's a lot to uh digest and quickly work out like what is worth listening to and what isn't. Here, hence why I've tried to put in so much uh supporting evidence is a confidence check. And what I mean by this is AI, the AI is going to tell me which parts of the output it feels most confident on and which parts it might

want a human to review. This just basically my way of like getting it to give the first pass and flagging to me which parts might be the the most uh important for me to look at or the ones that it's less least confident in. So it's most likely that I will have adjustments to make on it. If it's super super confident about one section because it come purely from uh the voice of customer then I'm like okay I don't really spend too much time thinking about how we can change that but other sections that might

be the case. So it's going to tell me here the inference heavy calls are uh part of section five, section six, section uh three etc. So I can go back through these and I can amend them. And again this is I do not want you to take this as gospel. I want you to take this output. I want you to put it into your persona document and you're going to work off of it. is going to become the skeleton for what becomes your full persona analysis document which by the way should not be a static

document this should not be something that you make once and then you don't touch which happens in a lot of brands and I understand because this is like the boring unsexy work that no one wants to do but [snorts] this should be a living breathing document that gets added to regular that gets reviewed regularly because that's how you're going to have a real understanding real deep understanding of who your prospects are and how you're going to track everything as the new VOC comes in, new reviews, new ad comments, new Reddit threads, etc. Uh, you'll want

to make sure that they all make their way into this document. Um, below here, just some other sections, additional VOCC, additional data sources that might, uh, help um, and just bits about crossplay awareness. So I'm going to take this. I am going to paste it into my goo do and that is going to form the basis for the uh master persona document. So I'm going to call this one the reluctant returner. Take this. You can obviously format it properly. Um but this is the core of what we have here. And again as new things come

through we're going to add to this and we're going to build it out. I will then go and do that for every persona. So I'll go back to Parker and I'll say okay now do it for this one and now do it for this one. Again you can do it inside of Claude but I think the output is going to be way better inside of Parker for this specific exercise. So you should have by the end you know three to five tabs of personas each one going deep on exactly who that persona is. I would

usually not go less than three. I mean in this case I might do if I don't feel confident about this the the number three and number four. if you remember from earlier. Um, and I wouldn't normally go more than five because because then it becomes a lot to manage. If you do, make sure that you have uh your core personas understood. So, in the original prompt that we entered inside of Claude and Parker, I asked it to rank the uh personas based on their level of prominence, how often they come up in the voice of

customer. So, if you do that, in theory, you should have your biggest and most important personas at the top of this document. So it makes it nice and easy for everyone to understand. These are our primary personas. These are our secondary personas. These are our like expansion personas, etc., etc., etc. You can make this document as long or as short as you want. Um, ours can be, you know, 20, 30 pages. Just don't make it, you know, so long it's just, you know, not able to be read or understood by anyone else who's looking inside

of here. But, you know, it's really important to to make this as detailed as possible because the better you know your personas, the better you know your audience, the better ad you're going to be able to make for them. And remember, we also want to be topping up this document with things that we find in our own manual research. So, for example, I have searched Laura Gella into YouTube. And um as you can see here, I found a video Laura Gella baked foundation review wear and test. Let's go into the comments and see what we

can find. See if there's anything we can take for our uh persona analysis. And um what I'm looking for here is comments that are, you know, insightful. We can add to our persona analysis. comments have got a lot of likes. Um, so it's supported by other people. Uh, I I'm not too interested in people just saying this is great. I love it. I'm looking for more detail than that. Like I am 56 with already skin and I love LG's power foundation. So if that was a persona, that's something I'd go and take and I just

sprinkle that in to the voice of customer section um in the persona document. Do that for Tik Tok comments, YouTube comments, uh, comments on your competitor's video. There's a lot that you can learn by looking at uh reviews of your competitors on Tik Tok, Instagram, YouTube, etc. And again, because it's difficult to get that from a Claude or a Parker. You can get some really valuable insights that worth your time uh digging in and doing the boring manual work here. Let's now look at step three, which is creative consumption analysis, which is somewhat similar, except

what we're looking for this time is to understand how this person already consumes content in their organic feed. So, you might be thinking, how can I understand what content they're already consuming organically on their feed? Well, something that we like to do is search the pain points of that persona into a Tik Tok, into a YouTube and see what comes up, like what type of content does come up. So, what we can actually do is before we do that, we can go into Parker and say, I now want to reverse engineer the search terms that

my prospects are using or this persona is using uh to find content on social media. give me search terms that this prospect would use based on their fears, desires, etc. Uh that I can type straight into Tik Tok or YouTube. We can just ask Parker for it's come up with these or Claude. Um alternatively, you can go and search within yourself, but uh this is a really easy way to get a lot of search terms super quickly. So, best foundation for over 60 skin. I like that. So, let's go into Tik Tok and search. We're

just looking for what kind of content comes up. Where are what type of creators are being used? Uh is it shot high prod or low prod? Uh is there a ring light? Uh like what they doing? Where are they? You know what does the background look like? And I can already see there is a clear theme in pretty much all of these pieces of content really. It's a it's a woman uh in often her bathroom, sometimes in a makeup chair uh or sometimes in the kitchen with some kind of artificial lighting setup for the most

part talking straight to camera a lot of time without cuts just very raw and authentic. I mean, it makes sense. It's a makeup product, but you get the point. This is the kind of things that we're looking for. the type of content, like how it is set up, how it's put together, because then we want to reverse engineer something like this for our ads because that's how we blend in to stand out in the feed. So, I take a bunch of these links and put them straight into our uh document in in a new section

called creative consumption analysis. I also would do a summary of what I just said. So, all of the common themes, you can alternatively put it through Gemini or Parker and say, "Here are 10 videos that have gone viral or have done very well organically. spot the current current themes because I'm doing a person analysis. I want you to uh put this report together for me or you can just do it manually. I would probably just whisper flow it and do it manually and have that section inside of here just for you to understand exactly um

the kind of content that this persona is already consuming. And you're going to do that for each of your personas. Meaning that once you've done this, you should have the full persona dock with everything that you need to know about each of your core personas that is ready to be used in your creative strategy approach. Let me talk you through the process of how we actually use this information in practice to make ads. Step four is persona ranking. So, what we're going to do is we're going to take a look at all of our different

personas. I'm going to assume that we've done this whole process for all of our personas. We're going to take all of our different personas and we're going to weight them uh based on the importance of these different variables. So, I don't have a super fancy process for this. I actually keep it pretty simple. I just take a a super simple table. You know, you can do this however you want. The the principle is more important than the execution or like how you execute it here, I should say. I just ran them out of 10 for

each of these different variables. So, let's just say that, you know, this is [snorts] a seven, this is a five, four, two. Let me just go and fill out all of the rest of these with just random numbers for now. And this is going to tell me uh it's just going to give me average score. There's nothing there's nothing crazy about this. But the reason that this is important is that your persona shouldn't just be looking at what's working in the ad account or it shouldn't just be which personas are responsible for the most amount

of revenue to date. You should be considering these four factors, revenue, LTV, CAC, and how much of a strategic priority is for you in the decision of where to prioritize your creative for the next sprint that you're doing. In this hypothetical scenario, if I wanted to say run three personas for my next batch, I would choose number one, number two, and number uh number four, sorry. And I might say, you know, 45% of my content is going to be around this. 40% is going to be around this, and the remaining 15% is going to be

for this persona. This is really important because a lot of people ask me how you test with intention when it comes to ad creative. And I really do think it is a top-down approach from your personas. So when it comes to uh us planning our sprints for the uh for the week, for the month or whatever it is, this is actually the spreadsheet that we use to plan all of our content ahead of time and we will track each ad that we make on a persona by persona basis. So, we can say, "Oh, for the

next month, we are going to make x% on this persona, y% on this persona, and and z% on this person." I I just hate that I said z. That's really American of me. Um, and then we can we can actually plan that out for the next month in advance. What we also have inside of here is uh the hypothesis. So, one thing I found this to do for our strategy team is it's really forced to think about why we think this ad will work. Like, why are we testing this persona for this concept or this

format? like what is the what is the theory that we are testing and then we can close the loop on that uh in a moment. But I really would encourage you to start testing your ads on a persona by persona basis. Not just for like looking forward, but also to look back. You can produce reports inside of Parker where you can say, "Look at all of our concepts for the last 30 days or the last 90 days or whatever. Break it out by persona and uh then go and give me a report on top spending

or top performing personas so that you can use that for your next batch of analysis for the next sprint that comes around. So I would encourage having some version of this whether it be inside of a spreadsheet inside of a notion database. It doesn't really matter. What more so matters is that you're tracking ads by persona. You actually will be able to do this inside of Parker very soon via AI tagging um but that's not live yet. But it does track a lot of other things like awareness level and things we spoke about earlier. So

you can actually pull these reports already inside of Parker um and break them out by spend by variable but just not personally yet. It should be coming very very soon though. Once we've done this, what our media buyers will do is produce a report uh looking back at the sprint and saying, "Okay, this was our hypothesis. We did testing for persona X, Y, and Zed." Uh, and this is how it worked out. And that would that would be used to plan the next uh sprint. So, this is like a sample monthly report. I mean, I

can't actually show a real one because it would have sensitive client data. Um but you know it would just be an exact summary um what moved what scaled winners and scalers and then the persona signals which is what I'm talking about saying like this is what our hypothesis was this is what we tested this is what performed didn't perform I mean you know this this report is maybe eight slides long but it actually ends up being like 25 or 30 uh pages um in reality with all the different uh slides that that come off from

these different sections but just basically saying uh this is like what happened over the previous uh sprint, whether that be a week or a month or a quarter. Uh for us, you know, we do do weekly sprints, but we do these reports once every month, just cuz it's easier to manage rather than doing them every single week, and it allows us to plan a bit further ahead than one week. So, sometimes our personas will change week to week, but for the most part, we keep them top level month to month uh instead. And I would

encourage you to have someone on the creative team do this work because it's so easy as advertisers to be in like week to week mode. It's a Monday morning. I've got a blank slate. I've got to make 20 ads by Friday and that's all I'm focused on. And you just keep sprinting and sprinting and sprinting. You never actually come up for air and sit back and go, "Oh, what's actually happening inside the ad account? Let me step back and plan for the next week, for the next month, for the next quarter. How do I want

to allocate my creative by persona based on what's already happening and based on my voice of customer?" It's really important we get our strategist at Adrete to actually block time uh to go and do this work or our performance managers at AdCrete to block time to go and complete this work because it's the boring unsexy work that allows you to actually test with intention rather than just throwing spaghetti at the wall which is what so many brands are doing nowadays. And that basically is step five of the process allocating the creative uh against the persona

ranking and and continuing to do that for every batch that you have. What you'll also want to do when you do this process again is obviously you're not going to do this entire process every single month, but you want to come back and and update the document. Make sure that any new voice of customer gets added in there. If there are any new personas that you've identified, you can enter that uh that into Parker Claude. Say, "Here are my current personas. Here's my voice of customer." We'll look through my ad comments, reviews, etc., and see

if you can find any new potential personas that are worth promoting into our core personas or and our master personas document. Um so make sure you have an updated persona document. Then do the ranking allocation again and circulate the report or the recommendations to the team uh so that everyone is aware of the intention and the uh testing priority of each of these personas for the next sprint of creative. That is the process uh for the first part. Okay, I know that was quite a long and detailed uh process. The other one's actually a lot

simpler, just a review of the ad account. But like this is what I would encourage each and every one of you guys. There's not one business owner or or a team member watching this who cannot benefit from going deep on their personas regardless of whoever you are, what you sell, and who you sell to. So the second part of this is actually looking inside the ad account to see if you can find any underserved personas. This is how we have found so many unlocks for our clients. You can find winning ads by cross referencing your

existing personas versus the ad account to see if there are any personas that come up a lot in your customer reviews, ad comments, etc. that are not being fairly reflected uh in the ads that you are making. So, let me show you how we do this analys. It's actually very simple to do inside of Parker. Again, everything I'm showing you here today, you can do inside of claw. You not do not need Parker to go and execute this, but the outputs will be a lot better and it'll be a lot easier inside of Parker because

of the data that it has access to. This is a visual report that I love to produce inside of Parker for all of our clients because it really quickly gives you an idea for any underserved zones that you have inside of your ad account. I'm going to hide these personas because it is actually uh a live client account, which I obviously cannot show, but I can show you the report nonetheless. You can open this inside of the chat itself or you can open it as a HTML that you can share with uh your uh team

or with anyone. Uh so I can see here on the left hand side of this chart in blue there is a uh the ad spend um and the percentage of the ad account. On the right hand side here you can see the customer voice. So it's looked at post purchase surveys, reviews, ad comments etc etc etc. And we can see that for uh the second persona, it is responsible for 30% of the customer voice, but only 18% of the ad spend. So why is that? Is that because there's not enough ads for this persona? Is

that because we've made a lot of ads for this persona, but they haven't really worked? Is that because uh this is a persona that like they are absolute super fans and they are more likely to leave a review than not is actually not a representation of our full customer base. Those are things you have to look into. But I can say like looking at this information, I have been able to find a lot of underserved personas whereby I mean we had a situation similar to this with the client recently where they had one of their

personas was responsible for 50% of their reviews but only 10% of the spend inside the ad account. So, we looked at that and we went, "Oh, if we're only making, you know, 10% of our ads around this persona, uh, but one in two reviews around this persona, that's probably something that's underserved." So, we started making a lot more ads for the persona. And now, this persona is working really, really well inside the ad account. Three of the top five ads. So, those are things that you want to pay attention to on a regular basis. Are

there personas from or angles from your uh voice of customer that are not being fairly reflected in the ad spend? It's not always going to be one to one. It's not going to be like, you know, all of like it should be identical to your voice of customer because there are certain uh things that can skew that data either way, but it it is interesting information that you can uh learn a lot from, especially if there are bigger discrepancies in this data. And these reports in Barker are great, by the way. Like I can respond

to this and say, "Now I want you to send me this uh in Slack every month uh on on the first for example, or I want you to send this to me every week by email." You can literally prompt these prompt exactly what you want in the chat. Um, and you can get it delivered to you whenever you want it to be delivered to you. You can manage your uh different reports inside of here. You can pick from from any of our template reports that you want to set up or you can literally prompt your

own new report. I want you to give me this exact thing, whether it be a weekly performance report, a monthly ad account audit, visualization of of customer reviews or voice of customer. Anything you want in here, you can prompt it. You can then say how often you want it to be delivered to you on a cadence, Slack or email, and get it pushed to you in your team. So you can get these insights when you need them without having to go and do the work every single time yourself, which I found to be massively insightful

inside of Parker. And this is something I actually have set up on a cadence for a lot of our clients inside of Parker. So that process was obviously a lot quicker than the uh the first one, but equally as important to look at your ad account and see if you can find any underserved personas or angles as well as identifying, you know, what your core personas are and updating those on a regular basis. To recap, what I would recommend you doing every week, every month, every quarter, uh is examining your current person's performance in Parker,

see if there's any that underserved. Check your voice of customer for any new reviews, add comments, support, etc., and add those into the master document with a combination of uh your own research and the stuff that you find through AI. Add any new personas that you identify into the document. Complete the ranking and allocation exercise so you know what personas you're going to be hitting and at what proportion for all of your ads for the next sprint. And then if you if you want to, you can circulate the recommendations to the team. That's what we

do as an agency. But, you know, that's that's up to you. And that is pretty much the process, guys. So, if you found this valuable, drop a like, comment below what you want to see more of because I am really trying to be consistent on YouTube this year. There's a lot that I want to talk to you guys about and I want to hear what you guys want to see. If you want to try Parker, there's a link to the description below for you to go and try out today and um start making more winning

ads. So, thanks for watching and I will see you in the next