My Simple 4 Step Facebook Ad Framework That Made $10M

if you don't have a proven system for launching testing and scaling your ads you will always struggle to predictably and consistently scale over the long term yes you might get some quick small wins over a certain period of time but if you want to manage and scale big budgets you need a system that you can stick to and you can also provide to others so they can also use it for their own results as well and that's exactly what we have done so in this video I'm going to outline the 10 million dollar ad system

we have used to generate millions of dollars for our clients from Facebook ads now I know it looks a bit confusing on the screen right now because there's like org chats and stuff I guarantee that it's actually more simple as you than you think when we get into the detail which I'm going to go through right now so there's four parts to our system you can see here phase one is audience identification phase two is creative optimization then we move on to demographic dialing and then we go on to scaling and let's jump into number

one so audience identification so the whole goal of this is to find the best audiences for this offer slash business so we have free campaigns number one is a one campaign that targets basically the warm audiences nothing special there we have a second campaign that targets called interests and we have a Third campaign that targets look-alike audiences now the structure that we have found to work the best is having one interest or one lookalike audience per ad set so we don't stack anything on top of each other and we have free ads within one ad

set and across the campaign the ads the set of ads are identical so you can see here ad copy one add copy one add copy one add copy add copy one image one add copy one image two add copy one image free so there's actually a couple of things we're testing here the interests or the audiences we're testing right because the whole point of this is to identify the winning audiences but we're also testing the images we're just using one piece of AD copy so the variable the main variable we're testing is the interests or

the audiences or the lookalikes but we'll also at the ad level testing three different images because why not right because you can there's not two variables here there's just one the image so at the end of this testing phase the main again the main issue or the main goal is to find the interests that work best but then also to find what image also works best but the main focus is the audience identification once we have run that test and you can see just some details here you can you can stop the video and look

at the notes on the left hand side in terms of like testing phases and budgets etc etc but after we have identified the audiences that work the best we can move on to phase number two which is creative optimization so you can see here and this should be campaign number three you can see here we've got three new campaigns and the whole goal of phase two is to find the creatives that work the best so in this example we found 10 winning audiences you can see here the actual ad set level here is one to

ten you might not have 10 winning audiences you might have more obviously I just did 10 for a round number but depending on how good the phase one performed you might have more you might have less now as the fair suggests we're looking to optimize the creatives so we're going to try and find the best creatives now you can see in phase one we only tested one piece of AD copy which obviously isn't good enough we need to test more ad copy so that's going to be campaigns number one goal is to find the best

piece of AD copy so we're going to choose the best image so we've tested three different images in phase one one two and three whichever one perform the best we're going to take that image and we're going to put that image now with three pieces of copy add copy one from phase one but then we're also going to create ad copy two and add copy free and put that into an ad set so you can see here we've still got one ad set one audience one target interest per ad set free ads per ad set

but right now we've got the same image and we've got three different pieces of copy and we're going to run this test to try and find obviously the only variable is the copy the image and the interests I mean we're testing different interests but there's only one interest per ad set so the main variable we're trying to find figure out here is the ad copy campaign number two is we will test it with video to see what performs best this campaign one with the images or campaign two with the video so you can see here

we basically can just duplicate this campaign and we can change the image for the video everything else stays the same the audiences stay the same the free pieces of odd copy stay the same the only difference is that we're trading the image out for the video just to see if for this offer for this business if video works better than static images campaign number three is we're going to be split testing more images so we've obviously seen from ad copy one over here we found one image that works the best now this could be let's

just say it's for a personal brand right this is an online coach image one could be of them speaking on stage image two could be them home with their family and then number three could be like just one by themselves whereas like a selfie style face to camera right the idea is that you test different types of images there's no point here testing someone's them stood on stage and then a different angle of them stood on stage and then a different angle of them stood on stage right because they're too similar we want to test

contrasts here so we could do Authority shot of them stood on stage we could do like family lifestyle um with them with their family and then number three could just be like a lifestyle shot of just them by themselves like face to camera selfie style right more natural now let's say for example that the family lifestyle photo perform the best so you can see here the same piece of AD copy but want the different images so let's say image two worked the best so what we're going to do now in campaign free is we're going

to take that best performing image you can see it here image and we're going to create variants so it was a lifestyle photo of them in the family that worked the best it worked better than the selfie style and it worked better than the authority shot so now we know that the family lifestyle photo performs the best we can split test variants of that photo so maybe we pick maybe this was a family photo of them indoors maybe we pick a family photo of them outside and then we pick a family photo of I don't

know them out for dinner or them at the beach or them doing like some form of fun activity right it's obviously going to be dependent on what type of creatives you can get but that's the strategy is in this campaign to be optimizing the the winning image and see which ones work best and create variants of that image to see which one hits home and we're testing it with ad copy one the same ad copy from the original campaign so we can see which one works best so at this time we're essentially going to figure

out which piece of AD copy is performing the best we're gonna decide which image or video like this image or video work the best with this offer and then if it is image we're going to see which variants of the image to hyper optimize works the best and at this point we're ready to dial in the demographics so the whole point of this goal by now we have already got the audiences that work best we've already got the creatives that work best so now we just need to dial in the demographics so demographics is essentially

age range genders countries and then also we have placements in here as well which I understand isn't a demographic but you know it's just part of the process so what we want to do now is analyze the previous data from all of the different campaigns and let's say in this campaign number three we analyze the data and we saw that a lot of the cheap clicks and the clip cheat leads and the cheap results were coming from a certain age range let's just say it was 45 to 65 then what we would do is duplicate

say the best performing five ad sets from this campaign into a new campaign everything would stay exactly the same except we would just Target that specific age range and let's just say in this campaign this video campaign that we looked at the data and females were responding better to this art and this campaign than males so we'd pick the five best performing interests sorry audiences or just whichever audiences were within kpi so out of the 10 maybe it was two or maybe it was 10 of them right just depends on what the data says but

if females will perform responding better then we would duplicate the winners and we'd put them into a gender Campaign which was targeting female only obviously if males responded better then you would do a male only campaign what we can also do as well and this depends on the offer if you can only Market to one country let's just say you know the the company only ships and serves people in the US then obviously you can't do a campaign targeting a different country but let's just say for example there's a couple of things what you could

do maybe initially you targeted multiple countries so let's just say you did the top four which is U.S Canada UK and Australia what you could do then is again you've looked at the data and let's say in this first campaign that people from the US responded best to the ads you could take the winning ad sets you would duplicate it into a new campaign and test us only or alternatively if you only targeted the US in the beginning but you have the ability to Target different countries then you could just do a random test of

okay so it worked for us maybe it will work for Canada so you would pick the best performing campaign the top performing ad sets within that campaign select them duplicate them into a new campaign and you could Target Canada so we've got four here but you could do multiple it just depends what the data tells you if there's if like there's a good age range for all of these campaigns and you could duplicate them all once and do an age range test so that would be free campaigns maybe females responded better to this campaign this

campaign but males responded better to this campaign when you duplicate this female duplicate this female duplicate this to male right so there's three more so you could actually have countless ones it just depends on what the data tells you right it's all data-driven decision making and then another final one we like to do is placement so again analyze the data if there's anything within these campaigns that say Hey Instagram feed people from the Instagram feed responded better to campaign number two pick best performing ad sets duplicate it into a new campaign and Target IG feed

only the same with Facebook feed Instagram stories etc etc you just got to analyze the data from all these campaigns and then choose which one works the best and that is demographic dialing it's basically taking what's working and then really dialing in so we're targeting that specific segment of the audience so at this point we know what audiences work best we have the best creative combinations and now we know what demographics and placements are working the best now we're ready to scale and this parts are super simple all you do is look up you can

look at all the campaigns but specifically these ones because these the ones from phase three should be mostly dialed in right now but if you've got campaigns earlier on I mean AD sets earlier on that in campaigns earlier on that are killing it then you can still scale some of those but essentially what you would do in this part is look at any of these campaigns and see which one is working the best let's just say for example campaign and again you can have more than four you could have you know 20 of these campaigns

in phase three depends on the data let's just say campaign number one was working really well what you want to do is pick the basically the ad sets that are within kpi so out of the five that you've got here let's just say four of them was within kpi you could pick those four and you could duplicate it typically what we like to do is three to four ad sets in a scaling campaign so you look at your campaign and you would pick the top two to three to four ad sets and duplicate it into

a CBO so let's just say this one was working best and audience one two and three was working the best we would duplicate it and put it into a CBO you can see it says CBO here right now we've gone from this is the first time we've gone from ad set budget to campaign budget now you want to duplicate it you want to keep everything on and then you want to let it run the budget you can see the formula here is the CPA times 50 divided by 7. so real quick how the algorithm works

the whole goal here from phase one to three is to figure out what's working these ad sets will eventually die why because most likely they aren't going to exit the learning phase if you want to scale over the long term you need your ads to exit the learning phase the way you exit the learning phase is having an individual campaign or an individual ad set get more than 50 conversions sorry get the ad set more than 50 conversions in a seven day period now most of these ad sets are at a five to Fifteen dollar

a day budget and depending on the campaign most likely they aren't getting over 50 conversions as in a seven day period because if it's got a five dollar a day budget seven day period means it's going to spend 35 that means that it's going to have to have conversions for less than a dollar if it's going to get 50 conversions after spending thirty five dollars so you can see that it's unrealistic for these ad sets to get out of the learning phase so the whole point of the CBO is to find what's working the best

put it in your CBO with a larger budget so then it feeds the ad sets a big enough budget for it to exit the learning phase so again the calculation to get out the learning phase is 50 conversions in a seven day period so the budget that we set these cbo's at is the average CPA of the campaign Times by 50 divided by seven so you're setting it at the daily budget that should get 50 conversions over a week so say for example that your average CPA here from one of these campaigns was let's just

say it was ten dollars so what you would do is you would do ten dollars which is the average CPA you would Times by 50 which is 500 bucks then you would divide by seven and that number I'm not going to do the math in my head but it would be 500 divided by 7 that would be the daily the minimum daily budget typically what I like to do as a general rule is my cbos don't really go any lower than 100 a day even maybe if this calculation comes up to like 75 dollars a

day I'd still probably just put it as a hundred just because then you're giving the algorithm a bit more room to work because realistically you know you're not going to get the exact CPA that you were getting in phase three it's probably a tiny bit higher because we're obviously trying to scale so that's like a general rule CPA times 50 divided by 7 and then probably just round up it's kind of like the loose rule and that's what we do and again you can have multiple of these campaign ones right because if you if you've

got free ad sets in campaign free that's working you make a CBO out of it if you've got four ad sets in campaign one that are working you make a CBO and you can even go back to these previous campaigns and if they're getting a good if in campaign two here there's four ad sets that are working then I would duplicate those four into a CBO if they're continuing to work then let's just try and scale them now as a general rule that's what we do we scale into a CBO what we can also do

for hyperscale is to do some cost bidding so what we like to do is cost cap one cost cap two and a row as so cost cap is basically putting a cop on the cost a cap on the cost sorry so for example we can if our average CPA was a hundred dollars from the other campaigns then we would do a cost cup that says we only want to pay a hundred dollars so that campaign's only going to spend money if we can get a conversion for a hundred dollars right and again this isn't a

video on Cost Cuts or anything like that it's more of the ad structure so I'm sure there's more videos on this channel around cost cut structure but that's essentially what we do we do a CBO we do a cost cut one we do a cost cup two and then we do a row as which is essentially the same thing but it's looking at the row as number instead of the cost per conversion right so maybe we'd say we want a two rowers and that is it that is our four-part scaling system now there is actually

another phase after this where we basically hyper optimize these cbos but it's not 100 necessary you can very easily use this out structure to go from zero from 100 from 500 a day to spending five thousand dollars a day very very quickly following this four-step framework