
Loading summary
Ralph Burns
Hello and welcome to the Perpetual Traffic podcast. This is your host, Ralph Burns, the founder and CEO of Tier 11. And today's show is all about Meta Andromeda. If you don't know what that is, well, you're not alone, because Meta surreptitiously rolled out an update to their AI algorithm, which is one of the most massive changes. I'm not being hyperbolic here or just trying to use 25 cent words for no apparent reason, but this is the biggest change I've seen in Meta since I've been doing Meta ads. And we were the first Facebook ads agency back when Meta was Facebook way back when in 2012, 2013. And I've never seen anything like this. And this is the reason why we're doing a whole series of shows on this. Because you as an advertiser, you as a director of marketing, need to understand that the way that you were doing things prior on Meta no longer works today. And today's show is all about that. And fortunately we have one of the most innovative, I think the most innovative media buyers, digital marketing strategists on the planet, John Moran, on staff here at tier 11. And he goes through this stuff that his job is basically to innovate. His job at tier 11 is to innovate and find new ways of getting results for our clients and ultimately helping them achieve their vision. But doing it through new innovative strategies and AI is evolving so much. In fact, we do an AI Discoveries call every two weeks at Tier 11, just so people within the organization can show what they've been doing. Because this stuff is changing so fast, which is frightening for some people. But if you're a marketer, this is one of the most exciting times that I think we've ever had because these algorithms, these platforms are becoming so much smarter and they will allow you as an advertiser or as a marketer to be able to get results with less work. Yes, we have been promising that for years and years, but now is the time where that is actually happening. And I'm going to share my screen here on an article that came out. Well, it was actually, it was March of this year. It's called AI Innovations in Meta's Ad Ranking Driving Advertiser Platform. We actually advertiser performance rather. And we actually had a call with our new Meta agency representative this past week. We talked about the Andromeda update and this big change in the algorithm.
John Moran
And.
Ralph Burns
Quite honestly, it was sort of news to him, which I was surprised by. And I think there's a reason for that is because when these platforms, I remember back when I was an SEO, Google would announce changes, sort of after the fact algorithm changes. So they didn't want people to hack the algorithm. And I used to use link wheels and all this sort of black hat stuff to get rankings in SEO when I was an affiliate and then obviously doing it for some of our agency, small budding agency customers way back when. It's almost like Meta released this without telling anybody on purpose because they are leveraging the power of AI. They are leveraging the power of GPUs, which, if you don't know what those are, the things that are transforming everything in computing, it's all about compute. And this article here, which I encourage you to read, what I usually do is I usually take the URL and throw it into this app called Instapaper and it reads it to me because this is the reason why doing so many of these shows here is because of the ad ranking platform. And all the performance that we're driving right now is really is allowing the algorithm to do its job. So there's four AI innovations in ads ranking. I just want to get through here very quickly. This Meta gem, which is this super brain, it's generated ads, recommendations, model or gem, they're putting acronyms on everything because they're innovating so fast they have to like spit out all these new names. Basically it's like having a super brain that can read an entire entire library of books in seconds, or can read the entire Internet in seconds, understanding the relationships between all the characters, remember every detail and connect that into an understanding of sequence of events. The point is what they're doing is they're reading their entire algorithm in split seconds and then using that learning in order to place your ads in the right way. And then in this article which I'm showing here, if you don't subscribe to our channel, you definitely should. Over@perpetualtraffic.com YouTube during this initial launch, they did a test where metareals actually increased conversions by up to 5%. There's a very cool sort of algorithmic image here which sort of shows how this works. You're basically throwing all of your creative into an ad set and allowing Meta to take your product and the content that you give it and show it to a user at whatever stage of the journey in which they're in. And John and I are going to be talking about that in today's show. This is purely an introduction to wrap your head around it because I think a lot of people really have not been able to wrap their heads around this as of yet because it's super complex. I'm explaining sort of more of the why behind it. The point is this giant library, which they also call the Meta lattice, is a way of being able to sort all of the data that they have on all of us that use this platform every single day and then show the right ad format, the right type of product, the right type of ad, the right type of, you name it, content in front of the right person at the right time in order to move them along your sort of virtual sales funnel. So what Meta has created is sort of this virtual funnel inside Ads Manager. And John will explain this a bit further. Finally, in this article they mentioned Meta Andromeda and this is the thing that we mentioned when we talked to our Meta rep this week. They didn't know what Meta Andromeda was. It is literally the way in which they deliver this into the algorithm so that it's almost like you have a personal concierge who knows your tastes and knows your customer's taste so well that they just don't understand, like they like your shoes, but they know exactly which types of shoes and which colors of shoes to show them at just the right time. And we're seeing this in today's show. So there's those three parts definitely go over and you can actually see this and see sort of how these Grace GPUs are being used within the context of this system. Getting probably, you know, a little bit too scientific for some of the people that are listening to the show here today, myself included. But in essence, it just continuously learns from itself and all of this is super deep AI compute stuff. But at the end of the day, and if we look over on our content diversification chart here, which John's going to talk about here today, these are the types of ads that these eight ads right here are the ones that you need to now push into an ad set. Create ads around your service, your digital product, especially in E commerce, in the content diversification strategy, which is what we're going to be talking about here on today's show, is all about John just using two or three of these, not even using the full panoply, so to speak. Another 25 cent word of all of this content diversification. And so he got this chart here from an internal conference that has not been released. And that's what we're talking about here today. On today's show, I'll actually show you one of our TikTok videos that we did a couple of weeks ago that this is probably our most viewed content piece because it reveals this content diversification strategy. So John will talk today about the us versus Them content diversification ad which is getting great results in addition to ugc, which is primarily the type of ad that he's been using in this one particular case study, as well as six or seven others which we run here at tier 11. And we're getting tremendous results for it. So definitely check that out. Check out our YouTube channel so you can see exactly what I'm talking about. And I'm not gonna use this as a lead magnet. So you download it and give your email and your name and your email address. Point is like, this is a new way of doing things here inside Meta. And we're really excited to be talking about it here today with John with yet another case study in which we're now diversifying using this content diversification strategy which allows you to take your user and find the best piece of content to show you. The content diversification strategy, which allows you to show your product or take your product and find a user at the exact stage in which they are most likely to buy. And that's why you're probably listening to this show, is to get people the right type of customer to buy your products and services and ultimately scale and grow your business. So without further ado, yes, that was a longer introduction, but I feel it's really necessary for you to all to understand this because this has been done surreptitiously by Meta. But we're super excited about it here and that's the reason for today's show. So take it away boys. Let's talk Meta. Andromeda, you're listening to Perpetual Traffic. Are you looking for new audiences for your business? Well, Snapchat is full of engaged users you just might be missing. And if it feels like your current audiences on all those other social platforms are tapped out, it might be time to diversify. Over to Snapchat. Here's why. 75% plus of 13 to 34 year olds in over 25 countries use Snapchat. 40% of Snapchatters aren't even on TikTok Daily. And it's not just gen zers. In fact, nearly one in four Snapchatters are over the age of 35. And 85% of Snapchatters discover new products and brands through social ads. And that's 17% higher than non users on the platform. This is an untapped platform. I don't know many people who are advertising on it right now, it's still blue ocean for you. Because with Snapchat, you have the potential to reach a hugely untapped audience and drive incremental growth for your business. So if your paid ad strategy could use a boost, it might be time to put Snap on your roadmap. Go to snapchat.com perpetualtraffic to learn more. Snapchat is now giving away thousands in free ad credits to new customers. Go to snapchat.com perpetualtraffic to find out if you're eligible.
Amy
A lot of people don't know what the Andromeda update is, by the way. I'm finding this, like, I'm getting confused looks when I talk about it on discovery calls because I'm basically doing all the sales right now with Amy. She's starting to take over. But the point is, just before we get into the stuff today, just like from your perception, what has happened? How has this changed? What has been the big change? All of that. And we call it Andromeda, but it's hard to find. Like if you Google it, it's kind of secretive.
John Moran
It seems like, you know, it's funny is it's weird. Meta did this extremely quietly. And I spoke. There's a VC firm I'm attached to that's actually, it's outside of the VC firm that I'm with now is actually from a. An old client slash partner that has joined a new VC firm. And long story short, we were talking to some of their teams because I'm brought in as a consultant and when I enter into a situation, I try not to be like, threatening. Like I do have a good name in the industry and I'm like, I'm just here to help. And so a lot of times I'll hear from teams that are like an ivory tower agency when they're like, I'm like, hey, do you guys know like if the account switched over to Andromeda? And they're like, hey, what's Andromeda? So I explain what Andromeda is.
Amy
Right?
John Moran
They're like, I is this even mentioned anywhere in in Meta? Like I think Meta should have done a, an announcement on it. And funny enough, like if we back in December 2, 2024 and I'm sharing screen, is it like I'm on my laptop? Sorry. So this is, I'm like, yeah, it's engineering at Meta Supercharging advanced plus automation with next level personalized ads retrieval engine. Like there's a lot of information about it out there. It is Hard to find, but it is real. So a lot of people will look at me like, oh, here's John trying to trick everybody and act all smart and everything. I'm like, it's live. It's just. It wasn't public. And I understand what they're trying to do is force advertisers to be good educators and engagers again, rather than just running sale ads. So what Andromeda is essentially creative targeting. And that is extremely hard to grasp because for the last decade at least, we dictated targeting. No one else did. The media buyer, the strategist, the person pressing the buttons inside of Google and Meta and et cetera. So Andromeda essentially takes AI and interprets what you're trying to portray in your ad as a value proposition or whatever it may be. And it says, thank you for this ad. I believe this person should see it. Now we get to give suggestions. We get to say, hey, we believe this ad should go there. And Meta's like, that's a good idea. Thank you. I'll take that into consideration. And they will go and do whatever they believe is best. So the Andromeda update of the ad retrieval engine, using AI to find its audiences, long story short, just means it's up to us to produce the best diversified content in order to achieve, like, the most perfect sales pitch and place that in front of Meta's AI so that it can go find the user for us. They have not published how, they have not published why, they have not published the best standard operating procedures. They do not publish the details because they do not want us reverse engineering. I have a direct line to the Meta engineers that are working with Andromeda through a client of mine. And so I've been able to pick the brain of the people that are developing this at a very high degree. And I do that once a week. So where what's funny is we can believe this or not, but I am trying to lead by example. And that's what we're doing today, is we're talking about the creative category test that I'm running and what this is and how this is helping us achieve our target, both in platform, top line, new subscribers, et cetera. Because there are a lot of deniers, naysayers, like, I've always been the tinfoil hat guy, and I will forever be the tinfoil hat guy, because most of the time I'm first. I am first to find, first test, first market, first to share. But it is more like a, yeah, but that seems hard new. And my Meta rep didn't tell me that. And that's okay. If we sort of have a choose your own path model as us advertisers, there's no college for this. So I think what we're trying to educate the industry for free. You're welcome. Is how we do this to the best of our ability and what models you can follow so that you can level up your game. And the people who choose to deny and kind of ignore this will be left behind, period. And that started last month. So I think that that's the biggest thing that we have to look at is creative is doing the targeting and it's up to you to produce the creative. And technical media buying didn't get harder or easier. It is just drastically changed. And it has gotten easier in terms of structure, way harder in terms of optimization. And that's what we're trying to clarify for everyone now.
Amy
Well, it was an interesting article on the Verge that came out a few weeks ago. Well, actually, it was back in late May, I believe. And they talked about. They did mention in not so many words the Andromeda update, but Zuckerberg's intent for this. And it's not just necessarily to make the job of the media buyer better, it's to get rid of the media buyer. So that was their interpretation of this. Now, we've been hearing this for years and years now. Finally, we're at a point where, I mean, they're getting closer to a certain degree. However, there is a human element in all of this that still is going to require us, but it's going to allow us. A lot of these changes are just going to allow the singular media buyer to be that much more effective. And from a time perspective, like, as an agency owner, I'm pretty excited about that. I told our director of ops. I'm like our VP of Ops. I'm like, I ain't hiring any more media buyers. We aren't really like, maybe in some specialties we will, but we can like double our size and not add any headcount. Literally at this point in time. Maybe like another growth strategist, maybe another cso, that kind of thing. So the point is, like, this is coming. And yes, we are giving away all this information here for free, but it's still foreign to everyone. I was like, huge PI Law practice yesterday. They're like, ralph, what's the biggest thing right now in digital marketing? I was like, andromeda. They're like, what? They didn't even know. And they spend 50 million a year on digital. And this is like their head of digital didn't even know. So it's coming, it's here. But it's also. It's working which is great. And that's why we're doing the show here today.
John Moran
Yeah. And it's funny is like there really isn't a lot that you will I guess I would say is a lot of people that I know learn through feeds and if no one is talking about this, people don't learn. I find that very odd. But it is the way that we just are training even how we're running ads. But even in here, like this is even like AI's innovation and Meta's ad ranking driving advertiser platform performance. They even show how you give bulk it sorts the giant library of your ads. How it actually is doing ad creative personalized ad retrieval locking users into ad set and showing them repetitive ads for ad candidates. I mean they explain how all this works. It's out there, it's public and it's been there for since March. This is not something that is like ah, John made up a new word just to confuse people. Like, no, it's in. In a better. In a bad sense. It's not my fault. You didn't know.
Amy
We told you.
John Moran
I know. On Friday we're like, here's what to do. And they're like, no, it's like, no.
Amy
I want to do it the old way. I want to like separate out my levels of traffic. Like the old ecom ad amplifier level one through five. No, it doesn't work that way anymore. Right.
John Moran
And so I'm just going to kind of dive right in if you're.
Amy
This is. I'm good.
John Moran
Yeah. And I hope that. And I'm going to stop sharing because I'm in an Airbnb wi Fi. So my apologies everyone, but this is. The data that you're coming to see is not my face, it's the screen.
Amy
I might shut off too for some reason. Everything's very janky today on my side. But anyway, the show must go on.
John Moran
Exactly. And I hope everyone really just wants to see the data. That's all they want. Then, then the good advertisers will be interested in this and the bad advertisers will not see my face to hate. So I'm happy here. So here's a test that we ran the last seven days and the top line results are the same ad spend with 70 grand week one and 70 grand week two. I mean it was 1.21% difference. There is no difference in ad Spend now we dropped our NCAC by 25%. So our cost per purchase went from 282 down to 213.
Amy
Oh boy.
John Moran
And yep. And this is. This right now is done purely with creative. And what we did, and this will be very simple for most users to understand is obviously we're using the Andromeda model of one ad set, one campaign for our objective that we're trying to achieve in this target. And we typically. And this is for. Since December, we have typically only ran UGC ads that worked very well up to a point where we're only. We're kind of stuck at quarter mil a month spend and we want to level up to at least a half mil spend. So we have these, we have UGCs. These are people's dogs. Obviously we're talking about pet health and wellness and immunity and digestive and everything. So we basically tell the story of all of our dogs and you can see there's just UGC. UGC, UGC, UGC, UDC. I mean they're just UGC everywhere.
Amy
Shot with an iPhone.
John Moran
What's that?
Amy
Shot with an iPhone. Like this is not high end creative, but like it's working obviously. Right.
John Moran
Again, it's like I don't really care that I'm using an iPhone rather than a DSLR for the picture of a warrior here. So that's what's nice is exactly. Warrior is cool dog. But we've typically, we're running UGC and what we notice is something like this. Spend down 74%, spend up 777 then down 56%, then spend 38 up then up to 90. I mean you're seeing bouncing around of UGC. And this makes sense with Andromeda. Why? Because this is all of the same ad type. The same ad type is simply going to cycle through variations of the creative. Because if UGC can do the targeting, not all UGC will be able to all do the targeting for all users. That is not possible. It is trying to say that again. Yes. All ads being only UGC will not use all UGC ads to put in ad spend to do all of the targeting for all of the users.
Amy
And UGC ain't gonna work on everybody is what you're saying.
John Moran
Yeah. And it cannot be the only thing that expands. So it will get you to a point. You can take a founder story and build a healthy $2,000 a day in good sales. Can you make it to 3,000? Maybe. Can you make it to 4,000? Maybe. It's your glass ceiling is set by you, the advertiser, based on how much creative diversification you're willing to place into the engine to find the users who are not resonating with ugc. And that is the ceiling that we have placed upon ourselves with content diversification. And that is right. Even in Meta's literature. So that sending a link to that.
Amy
Article, because I think it's a great. I haven't even actually seen that yet. Hopefully it'll send.
John Moran
Oh yep, I'll be able to send it over to you. And I have a few.
Amy
One. I got it. I got it.
John Moran
Cool. And so what this tells us to do is we look at the nine and I'm so sorry. I was trying to get it on my phone but it just did not work as I'm My stupid wi fi. But I try to get it as soon as I hopped on. But there are nine different types of content diversifications. We've shared it a few times before. It's us versus them. Founder story, face to camera, ugc. Then there's nine different categories. We've been using UGC for a year and it's actually been working really well until we found out, hey, that's a great content type. But what about everything else? We're like, well, that's a really good point. So I'm gonna. I'm working on a model that we will be sharing in the next coming weeks, which is what I believe to be the best model to move into. But I won't share it yet because it's not done. And I like to lead by example. Theories are a dime a dozen. Anybody can come up with an idea. But to prove it with scale and aspen to a hundred thousand dollars is where the people that walk the walk or talk the talk can also walk the walk. So I will share that once that's done and scaled. But we are doing it in this account and you're starting to see the initial results of that. And what I'll share with you is the beginning of the model. Now here's what I'm talking about. This is something called SID Delivers more. This is actually a us versus them. This is one of those content diversifications that Meta is asking us secretly to produce because it's only in like closed circles. We exposed what those 10 categories were. Remove the AI ad because that's just an optimization. There's nine main ones and this is one of those. This is our us versus them. R1 bottle will give you more active mushrooms, more white turmeric, more vitamin C, more milk this milk thistle at a cost of $65, where our competitors are going to have, you're gonna have to buy a turmeric bottle, a milk thistle, a vitamin C, an active mushroom, and individually it will cost you $195. So what this is trying to show is the us versus them. I wish I can get this bigger. My mom, a little laptop. Anyway, let's see if I can. Well, hopefully that gives you a good understanding of what that is.
Amy
Yeah, that'll be good enough.
John Moran
There we are. That looks a little better. So now you're looking at and say, okay, us versus them and them is the industry. Don't buy from anybody else in the industry. We've already made it easy for you to understand why you should buy from us rather than anyone else in the industry. This is us versus everyone else. We place that in the ad. And our normal cost per result in this campaign is $213. This is getting a 1:48. And we can even see the. As soon as we add it in, it's starting to scale 163% in its ad spend in the last seven days we added about 10 days ago. So this is going from $395 to $1,000 a week. The cost obviously was, was $79 because it was really good. But as you double and triple things, they start to go up. But it's leveling off at a level that is still like $65 cheaper than everything else in here. And it is increasing my cost required first time or it's increasing our new customers by 40% this week. So we see that the us versus them immediately doesn't bounce around spend it scales. Meta is giving the ad spend to this ad on its own. Why? Because it is a different piece of content. In our diversification, we went from ugc, now have UGC us versus Them, like, okay, so now we have us versus them. Now let's say what if we did a product feature, which is fascinating by the way.
Amy
Just pausing right there just to see the fact that, I mean, we were all talking about how Meta has invested so heavily into AI and obviously AI is the algorithm. It's really, it's the same thing. It's machine learning or however you want to do it. The point is like they recognize that this is a different piece of creative that can be additive to other individuals who may have already interacted with one of your other UGC ads. And so it's allocating more spend towards it. And obviously the results are what the results are here, which are Pretty astounding.
John Moran
Exactly. Now, two weeks before that, we added a different piece of content. It was a product feature ad. Now, in our repertoire, we have ugc, we have us versus them, we have product feature. And I wanted to wait two weeks to share this because we've already established that UGC ads will bounce around and spend, which is a media buyer's nightmare. When things are going up and down 50 and 100%.
Amy
We're playing whack a mole in the old model.
John Moran
You're playing, you're playing. Yeah. And I hope it did well this week. Everyone is your optimization. Let's cross our fingers. That's what we're hoping for. And that is a bad way to scale.
Amy
Right.
John Moran
So two weeks ago on the, on July 4th, we added this ad that was a product feature ad. It is our immune defense that works vet strength formula. And we basically didn't take it. Us versus them. We just said us.
Ralph Burns
Hey, you know, when I was first a consultant actually doing the stuff that we're doing right now in Tier 11, one of the first tools that I learned how to use was from a company called Unbounce. And they are now a sponsor of Perpetual Traffic. And the reason is, is that their landing pages and how co you can create those landing pages without having to consult your designer, your developer. With drag and drop builders now built in AI copywriting, it's even better than when it was 10 years ago when I first started using it on my own to create my very first landing pages. These guys are absolutely amazing. They've got conversion optimized templates giving you everything you need to launch your pages on your own without developers. In fact, Unbounce is the leading landing page platform for building, testing and optimizing high converting pages. Powered by data from over 2 billion conversions. That is 2 billion conversions with a B. That means they know what converts. So if you want to Convert more customers one platform and launch pages fast, Unbounce is offering PT listeners a special offer. They are giving you the PT listeners 10% off when you enter coupon code PT10OFF over at unbounce.com forward/pt so head on over to unbounce.com forward/pt. Enter code PT10OFF and cash in. Today, convert more customers with one platform. Launch pages fast. You shouldn't have to wait for your designers and your developers to build and test your landing pages. Get started with Unbounce today.
John Moran
So this is us and this is our product and we have a slideshow of the active ingredients. We have all of the information about this product. This is our first test. What if we just feature one of our products. We added that two weeks ago. And look at the bouncing of ad spend. It has changed 18 bucks 3602 to 3583. Now this is not bouncing around between different us versus them. Why? Or sorry, between different product highlights. Why? Because it's the only one that's there. So it has stable spend. Now what's the result of this? Well, we went from 133 compared to 282. Now it's down to 100 bucks and it is about a little bit less than half the cost acquired first time customer with stable spends. And optimizing the cpa actually optimizing the NCAT because this is first click happy imports on nc, which is new customers.
Amy
Exactly.
John Moran
And this ad here is now delivering us a fantastic result where we're like.
Amy
100,000 impressions versus some of the other ones. Anyway, keep going.
John Moran
Yeah, and that's good. And so we say, okay, excellent. I want immune new customers. So it says we have 45 sales and 35 of them are new customers. And those 35 are immune new customers. Fantastic. It's exactly what I want you to do. I can still get the 11 returning customers. That will never stop. Meta will always attempt, try push and figure out how to go warm. First click cap imports will shift it back into the right direction, even against its will. We have bonked the AI over the head with a club and we're dragging it to the area we want it to go. And what that is doing is providing us with a reduction of 30% in new customer cost per acquisition for the same exact cost that we have been spending the week before. And we are optimizing this for even a reduction of 23% week over week. So this is how introducing a high quality content diversification strategy into the Andromeda is allowing us to reach more users at a better cost per acquisition because we've given it more ammunition in its repertoire to go find the users we want that we're not resonating with just a picture of a dog. Right, Brilliant. And so it's.
Amy
And you've really only content diversified into like now there is three total out of the 10 that we know that you intercepted in. You know that we have shared here a few times, but we know this and you don't out there. But anyway, we'll be talking about this a little bit more. The point is like the algorithm that just, it shows, it's just so smart. It literally knows More about you than the NSA knows about you. If you are in this particular niche, if you are, Am I actively looking, Do I have a dog? Where in the middle. It knows how deep you are in the funnel by sort of creating that for you and by diversifying creative, say, oh, well, maybe they'll actually do. This person, John, who is thinking about it, has seen one of my other UGC ads. Maybe what I can get them to convert is this new ad that's clearly different. It's an us versus them. Or maybe it's just a feature benefit, one. So the algorithm just knows. And this is the difference between how we used to do things. In the past, we would shut off all the crappy ads. Now, I did notice you had a few of the ads that had the higher cpa, the higher ncx. Some of them were turned off. Okay, but the point is like, this is a new way of media buying.
John Moran
It's a whole new way. Absolutely.
Amy
A whole new thing. And we've seen it all the way through. I remember like the only thing that you would do is you would create 30 ad sets under three campaigns, all kinds of different targeting, and then you would only put in like a max, maybe four or five ads in each ad set. That's all out the window. Completely out the window at this point. It is.
John Moran
And what's interesting about this too is when we're looking at what to do when, where and how and why, we're now in an area where it's like, okay, I understand the content diversification, I understand that I need to have some different variations of each category running at the same time. So we're testing and iterating. But the next thing I was asked typically is, well, how do I know when to shut things off, how do I know what's working, what's not, how do I trust the attribution or just what's the next step? Now you're adding in and I understand, but now how do I optimize and how do I shut off? We are seeing that because Meta is using a piece of creative for the targeting. If you have 2, 10, if you have too many of them, it will actually test and simply start doing that for you. So what we're seeing is that we have two kind of golden rules in now our removal of creative, that we have actually been able to prevent, plan and prepare for us to not be found in this whoopsie daisy type of scenario over here, which is inefficiencies. And I'll explain what I mean if I give 10 piece of UGC to a AD set. I have failed as soon as I start because, well, hey, wait a second.
Amy
You put in like 30 in here and scaled up to 250k per month and that's been. So you're doing well, right?
John Moran
Yes, until the plateau started to actually work down.
Amy
Okay.
John Moran
So we hit a ceiling and then we started to see this and things started to fall apart. Like we built the plane and it was flying in midair and then all of a sudden like the wing cracked and the windshield flew off. Yeah, we couldn't live up there for a second.
Amy
Everybody, like, I don't know what's going on.
John Moran
What's that rattling noise like? That's exactly what starts going on. So what we've figured out is that if we add 10, yes, things will work inefficiently if I just give it enough money. But if we would have known what we knew before, hindsight's obviously 20, 20, if we were to start off on a different foot where we wouldn't have to put 10 pieces in there, where things are just doing this like, okay, those are working. These are working now. Those are working now. These aren't working. And these are working. But why isn't that working? We were chasing our tail not understanding why. Now there's two golden rules that we have now that are following is if, if the CPA is too high above where we want it to be, and this is very important and Meta is starting to ignore slash reduce spend on that ad, we should shut it off. And it is becoming clear as day the reason why you don't want to do one or the other is because if you have an ad with high ad spend and high cpa, that's okay. That could be your founder story. Not a lot of people are going to be converting on the founder story. But if you have multiple founder stories or multiple UGCs, or you push into too hard into one category, you start to see something that looks like this and I'll share screen.
Amy
The lawn guy just showed up. So hopefully it wasn't too horrible showing your screen.
John Moran
Perfect. This. You can see that the amount spent on this one ad called Donna started off with $5000 per day and then very rapidly went from 5000 to 2500 to 1500 to 824 and then down to $73. Why? Well, the average CPA was 510. The best we got a conversion at was 250. The worst we got conversion at was $1266 for new customer. So it pushed in ad spend very hard. It found out that they are not converting and are way more expensive than the other UGCs in this campaign and decided to ignore it for other other ads. So what is interesting is now we give 2, 3 UGCs to Meta. Meta says, thank you very much. I will try to see what I can do with this. Compared to everything else, it is not working very well. I'm going to go ahead and shut that off for you. I'm going to start to ignore it for you.
Amy
What I used to do is just spend continuously.
John Moran
Exactly. But now with the creative doing the targeting, it would spend continuously sometimes because we're in an ABO environment and I still see a lot of people in the ABO environment saying, well, I'm going to force spend on it and you absolutely can, but if Meta doesn't want you to, you're going to have bad performance, but they will take your money. And that's the difference between letting AI optimize for who it needs to versus us ignoring it. And your mileage may vary. You may have to start off in an ABO environment, just at least get some data, whatever it may be. This isn't a golden rule to say never. The guidelines are. You shouldn't. So the guidelines are, if we are working in an environment where AI can leverage the content diversification for the individual funnels that each one of those pieces of categories it needs to go into, it will optimize much better. And this is something that I have floating in my brain that I have a very tough time explaining. But it does make sense when I start to kind of share this more in a situation like this. So, for example, let's say we have an us versus them ad, that us versus them is middle funnel. Right. That is like a middle funnel because it's like you already know about.
Amy
I'm aware of it. I'm like, not really. Yeah, I'll check out the competition. Maybe, like, it depends on how big my pain is or my dog's pain in that particular example. Yeah.
John Moran
And it's like, hey, my dog is getting sick and has valley fever and is susceptible to a lot of issues and has some issues with allergies. I really want to support their immune. Okay, so are you instantly going to go buy a Pet Wellness Direct product? No, you're, you know that you want an immune product, but you're still trying to figure out which one is the best. Now, if you have never heard of Pet Wellness Direct and you are looking to find out why you want to choose a certain brand or product versus another brand. You could have gotten there without ever hearing about us. So you could be looking at pet immune products on Google. You could be posting in Reddit, you could be asking a Facebook community, whatever it may be. If meta can identify that your online activity is involved in a immune product for dogs, they may not show you the founder story. They may show you the us versus them first. That's what's important is there are people who are ready to buy an umbrella today. Absolutely. So your shopping ad will convert a person wanting an umbrella. There also is a person standing out the rain saying, man, I really wish I wasn't wet today.
Amy
There's this thing, those two people exist, right? It's like raincoat or umbrella. I'm not really sure, maybe don't. They don't know the umbrella exists. So.
John Moran
Right. And so I'm using funny examples to prove a point. But there are people at the top of the funnel that you can market today that never heard about you. There are people in the middle of the funnel that would buy from you that I've never heard from you. There are people at the bottom of the funnel that need a product like yours immediately that have also never heard of you. So each ad that you add as your content diversification opens up an audience that you can target immediately without having to push through the entirety of the funnel. We found our glass ceiling on ugc. What is ugc? Middle and bottom of the funnel. Choose our product because what we can do. And now a person's like, well, what the hell, who are you? I don't even know what the product is. I see that you have a happy puppy, but I don't understand why. So that's where things like our full contact diversification will educate that user. But we can only grow so much by telling everybody, as soon as you're at this stage of your buyer's journey, we're the best. We just missed everyone before that buyer's journey stage.
Amy
And what was always interesting, whenever you use website conversions as your objective in meta ads, and this is going back to the Facebook days, is like we would always say, all right, your look alike audience is maybe 2 million people in the U.S. but there's a certain subset of buyers that are converters that are ready to buy like they're in market. And so what would happen is you would. This would be so common and this does not happen now, is that you would get. You would put your ads together like old school, the old way, but then you would get this super high frequency and you're like, oh, wait A second, there's 2 million people in this. Why is it like I'm being shown to like the same 100,000 people? Because those are the people that are. Only the people that are in market are ready to buy right now.
John Moran
Yeah.
Amy
So our answer to that was, well, we need more top of funnel Creative and then try to force the algorithm and then Facebook would still show those ads to those same hundred thousand people exclusions. And then you're like, you shut off the campaign, you start again and then the same thing happens and you're constantly chasing your tail. It's a whole new world.
John Moran
Absolutely. And what Meta basically said is we when you say, like, man, I really wish I had an umbrella. And then Meta's like, hey, have you ever seen my umbrella Umbrella that we have for sale? And you're like.
Amy
Was listening to my phone. Everybody that now they're listening. I'm like, yeah, it's great. I'm a marketer. It's awesome.
John Moran
Yeah, that's what's funny is it's like that's a tinfoil of hat thing. But I like to. I call conspiracy theories spoiler alerts. Like, we know that they know.
Amy
What I always say is, I say, well, Meta's not listening. They're just buying the data from the company that has the. Has the speaker on your iPhone and. Or they're buying it from Apple and they're not telling anybody. So. Yeah, but they still are listening 100%.
John Moran
So what's interesting about this is the content diversification allows Meta to take your product and find a user at any stage of the journey. That if you think about If I have eight powerful. Right. If I just simplify this. My UGC gave me 1/9 of my audience.
Amy
Yeah, probably. Yeah, right, right, exactly.
John Moran
My 1/9th of my audience got me to a quarter million to spend a month. What happens when I add eight more?
Amy
You didn't even change your time. Is this open targeting?
John Moran
I forget now I have no targeting, no exclusions. I have nothing. I literally just said thank you for your default settings. I'm gonna focus on creative.
Amy
Awesome. So cool. This is like one of the best ongoing case studies. I'm excited. I got that like the hair on the back of my neck and my. Well on my back. Right, that's outside.
John Moran
This mole is all tingling right now.
Amy
Anything else before we get to questions?
John Moran
Because the one thing I wanted to make mention just in the industry and I think that this is everyone and I don't even want everyone to knee jerk reaction. I want everyone to reflect on this question. We know Google goes warm. Google has become a harder network to prospect. Demand gen we've proven does go warm. Performance max we proven goes warm. If you don't believe us please just sign off this YouTube live like it's we're not for you. We've proven this for months but what that means is that we it's our job as advertisers to build an omnichannel full funnel if you do focus on meta because meta is a better, cheaper, faster prospecting engine than Google when comparing spend levels of CPMs. It just is, period. If we start here and we can replicate our success in demand gen now we have one plus one equals three not which one's better. So yes I'm taking approach of meta first but there is also a Google element that is going to follow and we have a case study where we did three different products in a meta campaign. I launched a demand gen with those same three products and what was working so well in Meta literally had 10 times the performance in demand gen because of that. So there is going to be a trickle down effect as in like the Google guys focusing on meta. This is how to structure your omnichannel flow and why everything ties together together and efficiency is key, not versus is key. So I want to make mention that yes we're focusing on meta at the moment right now because this is our largest scale platform we have at our disposal across all platforms and that is what is the tip of the spear that everything and an email back there going woo following this is meta is going to go, we have meta to Google to direct to organic to email to return traffic to return customers. All of these follow up the larger prospecting engine which is meta and if you screw this up your rest of your funnel will be inefficient.
Amy
Yeah, absolutely. And that's like we're just, we're really scratching the surface on this which is just so much fun because it's like there's so much more scale and Runway now with this Andromeda update and it's easier for a media buyer to manage. It just is if it's done the right way, if you have the patience. We've run into that before because time some products take 28 days, some take seven days, some take 90 days. So there is that and the algorithm and the campaigns have to understand like who you're going to target like when you first start. So there's a ramp up period Especially if you're switching all this over to Advantage plus Sales Automated. There is that. But this is the future of it all.
John Moran
Yeah. And I do have to say that in the industry, we're in a situation where everyone's ego has been reset. No one is a good meta advertiser anymore. Everything you were good at is gone. And you have to check your ego at the door and say, I need to become a student of the engine again. Because that's the only thing I've dedicated my life to, is learning. And that's why I typically find things first. So it's almost like, hey, what? You knew? Give it a nice hug, kiss goodbye, mourn the death, and we need to move on. That's what. That's where we are as media buyers. It's not that the death of the media buyer, it's that the death of the media buyer that cannot adapt.
Amy
It couldn't be. You know, that is the absolute truth. And here's the thing about this, too, before we get to questions, is like, you're kind of known as the guy who, like, hacks the algorithm. Like, you're not doing that here. You're not coming up with some crazy tinfoil hat scheme. You're actually doing what Beta wants you to do. Like, all the old way of doing things. Like watch a video of me explaining Ecom ad amplifier from, like, five years ago. That is so outdated. And it made us millions and millions of dollars. Granted. However, Throw it out doesn't work anymore.
John Moran
Don't fall in love with the way you make the money. Exactly.
Amy
Absolutely. And ego is. Well, ego is the biggest obstacle.
John Moran
Yeah. It's like I'm. We're talking to the guy that is right now. I was the one that basically says, hey, here's the nine content diversification strategies. Two weeks ago, I had nothing but ugc. We are building that like we're building other content. But I was the ignorant person two weeks ago because I didn't know I had to figure this out for myself. And so I wasn't just, like, forcing an old model. I had to learn to adapt. And I think we have to look at this as in, we're not hacking the algorithm and we are actually leveraging the algorithm. We're finding ways to pull the biggest lever for the biggest result. That's our job now.
Ralph Burns
All right, super appreciate you listening here. On today's show, we'll continue to rebroadcast those sessions that I do privately with John as well as you can also listen to those or watch those live on the Ad Lab every Friday. It's sort of a supplemental piece of content that we put out every single week. That's live. You can ask questions and that's over on all the tier 11 socials and LinkedIn, YouTube at tier 11 and you can get that@tier11.com YouTube and sign up for that. Make sure you don't miss a single episode of that. That's supplemental to what we do here in Perpetual Traffic, but alongside everything we do here, it really does give you a very deep dive into what's going on, the inner workings of how we're doing things and how we're innovating with AI, specifically with meta right now because it is so exciting. So anyway, wherever you listen to podcasts, make sure that you do leave us a rating and review. It allows us to get out to a larger audience here, do things the right way, teach you these new strategies. Do not live in the past, ladies and gentlemen. Live in today and with an eye towards the future. Keep it tuned right here in perpetual traffic every single week and we'll try and do that for you. So on behalf of my amazing co host, Lauren Heath Petrulo, who could not make it here today. Till next show, see ya. You've been listening to Perpetual Traffic.
Episode: The 8 Essential Ad Types to KILL IT on the New Meta Ads Andromeda Update
Release Date: August 1, 2025
Host: Ralph Burns
Featured Guest: John Moran, Innovative Media Buyer and Digital Marketing Strategist at Tier 11
Ralph Burns opens the episode by highlighting a significant shift in Meta's advertising landscape. He introduces the Andromeda update, describing it as the most substantial change to Meta's AI algorithm since he began managing Meta ads around 2012-2013.
"[00:00] Ralph Burns: ...this is the biggest change I've seen in Meta since I've been doing Meta ads."
Burns emphasizes the critical need for advertisers, marketing directors, and CMOs to adapt to this update, as previous Meta advertising strategies are now obsolete.
John Moran delves into the technical aspects of the Andromeda update. He explains that Meta has discreetly integrated advanced AI and GPU-powered computing into their ad ranking system, significantly enhancing ad delivery and targeting precision.
"[12:06] John Moran: ...Andromeda is essentially creative targeting. And that is extremely hard to grasp because for the last decade at least, we dictated targeting."
He introduces the concept of the "Meta lattice," a vast data sorting mechanism that allows Meta to deliver highly personalized ads by understanding user behaviors and relationships at unprecedented speeds.
"[02:49] Ralph Burns: ...like having a super brain that can read an entire entire library of books in seconds..."
John points out that despite the profound changes, Meta has kept the Andromeda update under wraps, with little public communication or guidance, necessitating proactive adaptation from advertisers.
Previously reliant on User-Generated Content (UGC), Tier 11 encountered limitations in scaling and increasing cost-efficiency. The Andromeda update mandates a broader content diversification approach, incorporating multiple ad types to leverage Meta's enhanced AI capabilities fully.
"[34:07] Amy: ...these are the types of ads that these eight ads right here are the ones that you need to now push into an ad set."
John introduces the nine categories of content diversification recommended by Meta, including "Us vs. Them," "Founder Story," "Face to Camera," and "Product Feature," among others. This strategic variety ensures that Meta's AI can effectively target users at different stages of their purchasing journey.
John presents a detailed case study showcasing the impact of content diversification on ad performance:
Initial UGC Campaigns:
Introduction of "Us vs. Them" Ads:
"[27:49] John Moran: ...a 30% reduction in new customer cost per acquisition for the same exact cost that we have been spending the week before."
Addition of "Product Feature" Ads:
John underscores that diversifying ad content allows Meta's AI to better understand and target various audience segments, effectively breaking past the scalability ceiling imposed by UGC-only strategies.
The shift towards content diversification signifies a fundamental change in media buying. Advertisers must now focus on creating a variety of high-quality ad creatives to feed Meta's AI, which, in turn, optimizes ad delivery across different user segments.
"[48:12] John Moran: ...we are leveraging the algorithm. We're finding ways to pull the biggest lever for the biggest result."
John emphasizes the importance of being adaptable, stating that the role of the media buyer is evolving. Success now hinges on understanding and cooperating with AI-driven systems rather than attempting to manipulate them with outdated techniques.
"[34:07] Amy: A whole new thing. And we've seen it all the way through."
"[48:49] Amy: ...the death of the media buyer that cannot adapt."
Ralph and John discuss the industry's general lack of awareness regarding the Andromeda update. Many advertisers are either unaware or resistant to changing their established strategies.
"[16:51] Amy: ...the head of digital didn't even know."
John notes that without widespread discussion and education on these changes, many will lag behind, unable to harness the full potential of Meta's updated ad system. Tier 11 is positioning itself as a leader in navigating this transition, offering insights and strategies to stay ahead.
Beyond optimizing Meta ads, Ralph and John advocate for an omnichannel strategy where Meta serves as the primary prospecting engine. This approach integrates various platforms like Google, email, and organic channels to create a cohesive and efficient marketing funnel.
"[47:27] Amy: ...the algorithm and the campaigns have to understand like who you're going to target like when you first start."
They highlight that Meta's advanced targeting capabilities can significantly enhance the efficiency of the entire funnel, from initial prospecting to customer retention.
Ralph concludes by reiterating the necessity of adapting to Meta's Andromeda update. He encourages listeners to embrace the future of media buying by diversifying ad content and leveraging AI to optimize performance.
"[50:09] Ralph Burns: ...Do not live in the past, ladies and gentlemen. Live in today and with an eye towards the future."
John echoes this sentiment, emphasizing continuous learning and adaptation as critical for sustained success in the evolving digital advertising landscape.
This episode of Perpetual Traffic provides a comprehensive overview of Meta's Andromeda update and its profound implications for digital advertising. By embracing content diversification and adapting to AI-driven ad targeting, advertisers can unlock significant performance improvements and scale their campaigns more effectively. Ralph Burns and John Moran offer valuable insights and actionable strategies, positioning listeners to navigate and thrive in this new advertising paradigm.