Is CPMr The Hidden Meta Ads Metric That Predicts Account Collapse? With Phil Kiel.

You are wrong about the metrics that you are building your meta ads account to pursue. If you're just going after a row ass and cack and you are not thinking about reach, CPMR, everything you're doing is wrong. Your account is about to break. That is not really the perspective of Phil Kill, but I don't think anyway. But Phil has been putting out a whole bunch of information and so it's Cody Plavker and myself and Taylor Holiday at CPC, a whole bunch of people in the digital marketing ecosystem right now. I've been talking about this metric, CPMR, which is the cost of your reach of your ads. Very, very simply how many people you're reaching and how much it's costing you to reach those people. Some people see this as the key metric right now to be thinking about at least at least as a leading indicator of whether or not your ad account is about to really sink, maybe as more than that, maybe as the leading indicator of how incremental your ads actually are. Smart people are saying smart things about this with nuance and thoughtfulness. And I am trying to get us sorting through that conversation. So to do so today, I brought on Phil Kill, the director of paid social at Tycoon, digital, who has been writing a whole bunch about this. And I'm gonna ask Phil a bunch about sort of how he thinks about this issue, how you reach people with your ads, who is missing out on potential growth because they are not thinking about this, whose performance is suffering because they're not thinking about this. And most importantly, how to use CPMR as a metric to diagnose problems in your ad account and improve your performance. Phil has thought really, really hard about this. He's written about it. He's responded to Taylor Holiday and CPC and they're studying about it. And so we're just gonna get straight to the source with Phil Kill. Right now, you're gonna like this conversation a lot. If you're like a media buying walk trying to think carefully about how to leverage med ads to grow your business from a media buying perspective, let's get into it. Phil Kill from Tycoon, digital, all the way across in Liverpool. Let's do it. (upbeat music) Phil, what's up, man? Great to have you on here. Great to actually finally meet. We were talking like the first DM I have from you was from 2020 when I was at four by four hundred still trying to sort things out. You very kindly pointed out that I had a broken PDP I think. And here we are. Here we are six years later working on very different businesses. Thanks, Andrew. Yeah, I mean, longtime, listen to the first time, call off, I think that's still the line. I've been listening to you for a long time. So appreciate it. Yeah, man. And you're out in Liverpool in England. Of course, like everybody else, I think I associate Liverpool entirely with the Beatles. So yeah, there you go. Yeah, that's it. So let's just get right into it. You have been writing my CPMR. And what I would like to start with is for you to sort of summarize the big picture position that you have. As somebody running paid social for a bunch of eight, nine-figure e-commerce brands, CPMR is something you've been talking and thinking a lot about. I've read your stuff about this. And we're going to link to your Twitter account. Let me just tell everybody right now, follow the link to Phil's Twitter account. You'll see a whole bunch of stuff about this. Phil, if you want to send over the particular posts that maybe guide people will put those in the show notes as well. Follow up on some of the conversation here. But start by summarizing your position for me. And sort of what is CPMR? Why should people care about it? And sort of how are you using it in out of accounts at the big picture? But start by defining it. And then what's the baseline position that you have? So in Meta, you can see this metric has cost per 1,000 unique accounts reached. And that's a column that you can add by default. It isn't a custom metric. And it works out as CPM times frequency. And it is the metric that tells you how much it costs for you to reach ads. Which is, and CPM is what probably one of the first metrics that you learn about when you get into maybe buying. We are buying based on impressions. It would be good to buy based on CPA or Rhoas, but we're paying for impressions. And cost per reaches, I guess, the next version of that is because it's CPM times frequency. And I'll pause now. I'll let you talk first. Well, so basically it's just a way of saying, it's just showing you not only how many impressions you've served, but how many people you've actually reached. Right? People have been talking about frequency forever. And so the notion of CPMR is saying, are you actually reaching more people with your ads or not? And so that's the whole idea. Cost per 1,000 people's reached. So CPM, if it's an impression CPMR is people. And I have an account that I'm coaching right now that is a really nichey product. And it's sort of a great illustration of some of the things that I think you've talked about where they started to try to increase spend. And all that happened is their frequency went up. So they reach really didn't go up very much at all. And their performance suffered a whole bunch. And you see situations like that. And you go, like, OK, well, what do you do? What do you think about this? And my Phil Keel sirens went off in my alarm. And in my mind, when I saw this, I was like, I wonder what Phil would do here. Because it's all kinds of things. And the other thing I'll say is there's a bunch of approaches to this question. I don't think it's actually new that you need to reach more people with your ads. I think that's obvious in one respect. But there's a lot of questions around campaign structure and signal engineering and stuff like that and what it's going to do. So with that defined, Phil, can you say a little bit more about what you notice about that metric and how it tells you something about the performance of an ad account? Why does it matter? Because what does it do about this? Have we always tried to reach more people? Yeah. I think over the years, we've become more and more-- we were even more focused on the lagging indicators. Because this is performance marketing. This isn't marketing. This is performance marketing where we're spending to get a result that we can measure. And naturally, we can over focus on the most important metric. You've got profit, CPM-- profit, contribution margin, row R CPA. Let's just think about those and ignore everything else. And CPMR is essentially a signal. There are plenty of other signals in your ad account that you can use. And there's been arguments, discussions about click-through rate, measuring creative based on click-through rate. I feel like that was a little bit more popular a few years ago as a discussion. And I guess this is the next conversation around trying to understand what is happening based on signals over the business performance, row RAS profitability. So we really use it as a signal. And I really got into it a few years ago now because of a specific account-- specific business, maybe two and a half years ago-- which had the settle to fall into this problem. And I think one of the previous podcasts you talked about, there are certain types of businesses where this is more of an issue. And it was an attention business. So it was more of a single purchase business. They weren't doing very much product development. They were over-indexed on Facebook. They were advertising in the UK. So the population is 60 million, which is smaller than the US. So you can approach problems quicker than what you could do in a bigger territory. And so they had all this settle that allowed the account to fall into this problem of CPMR. They were also maybe around six or seven years old as a business. I don't think you've faced your problem straight away. And so that's really where I-- OK, what is happening? Why is business getting harder? And then the end result was CPMR. Yeah. OK, so let's start breaking this down a little bit more. So when you say the end result of CPMR, what showed you that? How did you discover that first? So was there a report that you looked at that-- or is there something you set up for people right away? That was like, OK, this show-- you're reaching less people with your ads. Yeah, and that multi-gay thing is that's really important because we could compare data per year as a starting point. And we had all of these metrics from the start of the funnel to the end of the funnel, spend impressions, reach, frequency, click-through rate, after-cart rate, CPM, ROW-AS. And we created a table over multiple years. And we looked, OK, so what's changing? Why is performance now harder to come by? And the metric that jumped out that was changing to the most was CPMR. And that resulted in spending more reaching less people. Yeah. Yeah. OK, so there's two ways of-- when I hear that, when I hear you're spending more of reaching less people, I don't hear that as a media buying problem. I hear that as a creative problem. And I'll say here, I have just released two episodes talking about this issue from a creative lens and a media buying lens both and talking about how I have thought about this issue in the past. And the TLDRs that I haven't thought about in the past, and yet I would say I have at the same time by proxy. And essentially saying, my approach to this is, yes, you need to reach more people in the path that doing that is creative. So when I hear you say-- when I hear you say they were reaching less people, one way of approaching that is we need to do something from a media buying perspective that's going to help them reach more people differently. And for less money. Another way of approaching that is by saying, that's because they have done poorly with continuing to generate a high volume of high quality, how the difference is creative. That is rewarded now more than ever in the current algorithmic and machine learning environment. And therefore, they probably haven't kept up with that. And that's why they're reaching less people. That's sort of the way that I would interpret that is that they probably didn't need to do anything differently with their media buying. They just needed to approach it from a creative lens. And in that respect, CPMR would actually be a leading indicator, yes, of a problem to some degree, but it wouldn't actually be the problem, if you know what I'm saying. Like it would, it would be that there's something else that's causal to it. Now, I don't know if you agree with that statement at all or not. And by the way, both of the episodes that I just referenced that I did before are linked in the show notes. So you can go back and get those. And while we're on the subject, please do subscribe wherever you're watching or listening because we're talking about deep, media buying stuff like this all the time. And if you like this conversation, if you're this far into it and you want to keep going, you will like my content in the future. You will. We do this kind of stuff all the time. Probably I'll have Phil back at some point. We'll talk about it. And so subscribe wherever you're doing that. Leave a comment as well. If you have thoughts on this issue, I would love to hear them. I read and interact with every comment there. So get back to my question. Phil, so when you see that sort of where does your mind go next? What is the actual problem? If more expensive reach is the problem, what's driving that problem? Do you see what I'm saying? Yes. A specific example, it wasn't, while it can be creative and it is creative more often now because of the changes with Andromeda and whatnot and the way people are using platforms. The problem for that specific example was new product development. And the length of time that they'd had been advertising in the UK. And the amount of investment that they put into the account over that time. So therefore they'd reached most of the population in the UK with a product was predominantly a single purchase product. And they had been very slack with new product development. So there wasn't an opportunity. There wasn't a reason for the ad account to reach different people. There wasn't, there also wasn't enough reason for the ad account to reach the same people again with a different, with a follow up product. And so while us as media buyers and marketers were focused on the account structure, creative those activities were actually fairly futile as solving this problem. That's the sort of the biggest example of CPMR as the signal from outside ad account problems, business problems. Interesting. Or business, the lack of business activity. Whereas nowadays, the signal is increasing CPMR, decreasing reach and it usually comes down to, there's a big list, there's campaign overlap, poor exclusions, a lack of organic activity, a lack of activity from other channels. And obviously a creative and messaging being too focused. Okay, so, so in that case, what did you do? It sounds like what you're saying is that you needed to tell the brand, hey, listen, you need to make some more products. And now, by the way, I actually, I think it's actually elite tier of media buying to tell the client what's true even when they don't want to hear it. Because what other agencies will do is they will say, oh, we can fix this with these, the reason that your CPMR, the reason that your broadcast sucks because these, these the idiots over here are doing the media buying wrong and the creative wrong. But if you give it to us, we'll solve it for you. And that's what the operator often wants to hear, right? Because then it's not their problem and it's not a hard, it's not a hard, it's not somebody else could just solve it magically, right? But what, it sounds like what you're saying is that in that case, the brand needed to hear, you need to develop more products. And I can actually think of a brand I've worked with where what you're describing sounds very similar to them. It sort of, it seems to me, reached their maxed-tam. And they were, it was just going to be really hard to reach more people without product development. So anyway, what did you tell them? Or what did you do? We kind of, I mean, nowadays I would be a lot more, a little bit, I was sort of free years, a free years younger, free years, more less experienced. Yeah, yeah, yeah. And CPMR wasn't as common of an issue. It seems like it's a, you know, I have, yeah, I don't know if it's a self-fulfilling prophecy. I have a lot more people coming to me now, where CPMR is the issue, is that because I'm looking for it or whatnot? Maybe that's a, you know, there's a deep question. But, yes, we talked about it. They probably haven't solved it yet as a business, I imagine. I'm not working on the account anymore, but I expect they haven't solved it. Yeah. Okay, so let's say you, let's actually, let's actually talk about something. You talked about account structure and it's being more of an issue and a drama on some of those things. Here's one of my big questions with this. Do you, do you think that this is a ghost in the machine problem? And what I mean is, essentially, is meta working against you. And this is one of the things I hear from Cody Ploughcar a lot about this, which is sort of the notion that meta has a really hard time getting outside of it's sort of like loops of people reach and that you actually have to fight meta here. You have to work against meta's default behaviors to increase CPR, CPR and therefore increase your performance. Is that the way you think of this problem or is or or not? I mean, I have a, I have a response to that, but, but, but I'm curious if you think this is like a meta, if this is a problem in meta, that's a broad answer. Yeah, sure. It impacts more, some businesses more than other others because of the, the makeup of those businesses of the product and the time and the, and the territory and whatnot. So there are some businesses where this is not an issue whatsoever. And those are the people who are probably on Twitter who are a little bit more like what you're talking about. I didn't need to even think about that because, because they are advertising a product in a very large territory that is has a huge address, a more, a stressful market. I think it is a bit of a feature and a bug because, yeah, frequency is important. We all use Apple Macbooks and iPhones. If our frequency of Apple was one, we likely wouldn't have purchased the product. So and I feel like you talked about this before where there is, there is a, there's a point at which performance can improve as frequency increases. And then you could pass that and performance starts to decline or it's diminishing returns. I think it's difficult for me to know where that point is. I think it has the impossible for me to know where that point is. Unless you're using my, my, your bidding and the spend starts to then drop. I have a recent hire a new growth strategist at, at AGF Growth who's been really great as an addition to our team. And he said to me recently, he said, man, this guy, yes, he's talking, this guy, yes, from your team, he's so awesome. Like where did you find him? And I said, I found him with more staffing, more staffing is the place where I found his incredible employee who does all this amazing work for us got recently promoted. Yes, came in as a media buying assistant in a sort of introductory media buying role at AGF Growth. He's gotten promoted and I'm just going to tell you he's going to get promoted again in the future because he is an absolute beast. It was a critical part of our team because more staffing helped us locate incredible talent in the Philippines that we have every expectation of the world of continuing to create more opportunity for because that's what more staffing does. More staffing was built on e-commerce businesses. They understand every element of the e-commerce ecosystem as well or better than anybody else because they literally have been staffing Filipinos in e-commerce, US based e-commerce business businesses for a long time. It was so effective for them. It was so transformative for them for their businesses that they went and started a staffing agency to help make it possible for more businesses. That's what you should do. The reality here is really simple, which is that your dollar goes a lot farther to get incredible talent in the Filipino market than it does in the US based and the US market. It's just that simple. You can get great talent added to your team with dbcommerce resumes who are going to work hard, who know what they're doing, who grew up speaking English. If you're worried about that, just about every Filipino is an English speaker pretty much natively. You can find incredible talent for people who are going to really contribute. If you're thinking about this only at the $5 an hour level, you're just missing out on the opportunity. There's just more to it than that. You can get really, really good talent throughout your organization. We've continued to staff our team with incredible Filipino talent for more staffing and their sister companies. We expect to do the same going forward. You should do the same thing. Whatever position you're hiring in your e-commerce business, go to more staffing.co/af. Get on a call. Just do this. Just talk with them about what the job description looks like. See what kind of resumes come back. I bet you will be awesome. They will find you that talent. They will pre-interview, pre-screen candidates. They'll give you just the best of the crop that they find. You don't feel sorting through hundreds of resumes or anything like that. You can get down to just the best candidates. I bet you will find great talent for the position you're hiring for. Go check it out today. More staffing.co/af. Yeah. I think you're right about the frequency thing and actually met his own reports on this when they've talked about this. Last I've seen, last I've seen is that if a customer sees the same ad more than once, literally the same piece of creative, the performance immediately declines. There is no good increase in frequency. If it's the same ad. I don't think that's true if there's multiple pieces of creative. But if it's the same ad multiple times. There is an element here to where this is one of the answers. I'm curious if you agree with this. One of the answers here is a high volume of high quality, highly diverse creative. That has to be in the answer somewhere to this problem. Yes. Yes. I like to think about it as what's the experience that the consumers going to have as they bounce between ads and campaigns in the account. You want them to have a, you want them to go through a story with the brand and the product and that is therefore different creatives rather than the same one over and over again. Is there anything else at the level of a media buying that you look at for this and you say, these are other things that we do that are really important so that we can make sure to minimize this as a problem in so far as it comes up. Like you're saying, like some accounts, this is just not an issue based on spend level tam, all those kinds of things. But when you see those issues, are there any questions? They're like media buying or let's say for all the people who are coming to you saying Phil we have a CPMR problem fix it Uh, are there any media buying things that you see people doing wrong consistently that you think is really hurting them? Yeah, it's it's the things that impact The individual metrics that make up CPMR so it's CPM which is a little bit less Uh, you can impact it less because it's a you know, it's a platform cost based on the market and the product and the you know the niche that you're in So it's more often than night frequency. So it's exclusions and exclusions are not consistent The exclusions that worked for your account two years ago May need to change you may need to increase your exclusions because You're spending more your time is shrinking your product development is is low Uh, so changing exclusions over time and then consolidation So the overlap between multiple campaigns and assets in an account will increase your frequency Increasing frequency increases CPMR And so yeah, go ahead and And it's very common nowadays that we see people having too many campaigns and assets for the level of their spend They they don't really don't have enough ads or too many ads. It's very often it's not very often that I'm saying you've got too many ads It's more often than I'm saying you've got too many campaigns and adsets But practically what does that look like there is it like They have 10 campaigns and five ads as in each campaign, but only three ads in each ad set that kind of thing Not necessarily the last point. So 10 campaigns five adsets in each one a ton of creatives And then they're spending two thousand dollars a day Got it So that's interesting. So they're actually too much creative and too much split out and when you consolidate you cc PMR go down It will do it will do but guaranteed because frequency will decrease And then your cpm r will decrease and if you're spending the same your reach will go up And you could actually spend less and reach the same number of people All right, so let's actually let's keep let's keep talking about this so So why does why do multiple campaigns drive up cpm r? Why can't why can't meta just deliver some and not the others and I guess Yeah, yeah, what is the mechanism for that problem? So each campaign will have it individual reach If each campaign reaches a thousand people and you have 10 campaigns your account reach will not be 10,000 Your count level reach will be 7,000 And at that point there's a 3000 overlap between those campaigns. That's that's that's low I ordered it in an account last week where the account overlap was like 80% So that the number at the bottom was 80% lower than the total reach when you want it to reach individual campaign And that is an issue that is the bug that you asked about earlier which is each campaign Is not going after a total or a total different pocket. It's that Venn diagram And that but those circles are overlapping a hell of a lot But you haven't just got two circles with a Venn diagram You've got 10 Yeah, and you're increasing spend across all of them. Yeah, so for that account where you see in 80% overlap your solution to that problem will be To just To create less campaigns in the ad account So you could literally take all I mean, I guess this is what I'm asking if you took all the same ads that were active in that account But just put them into less campaigns and I assume in cbo setups if you did that That that would decrease the overlap even even so even if there's let's say there's just as many ad sets But there's just less campaigns is that what you would do or would you actually kill a bunch of ads get less ad sets as well Yeah, but by having less campaigns you would likely have less ad sets So the ad sets are the real issue because that's where the the audience is set So yeah, you would have less ad sets and you know move into cbo is obviously important because then some will receive less spend and will have the smaller impact from the overall account And then essentially increasing exclusions at the same time That's interesting. Okay, so come try to understand the mechanism is why is why I'm asking about this more and part of it is that I think I buy media So differently than what you're describing that it's hard for me to imagine but But Not I don't think it sounds like I buy it so differently than what you do I think it sounds like I buy it really differently than what these audited accounts do but But It sounds like what you're saying is if you if you have a bunch of campaigns And I think your assumptions that these are auto bid campaigns. Yes. Yeah, more often. Yeah. Yeah, right. That's what I would that's what I assume So if you if you are auto bidding Across multiple different ads. That's multiple different campaigns or a bunch of a bio ads. That's whatever Then meta in each ad sets level is just gonna try to find the most likely person to buy right Off of that and it is forced to spend money across each of those things and so With no ability to rank the ads It's set itself because the assets are all split out in different campaigns or it's the LABO or whatever Each of those assets is just gonna find that same small group of people and say like okay, well like You know, it fills the person likely to buy. I'm gonna serve him ad a and ad b and ad c and ad d instead of Instead of in a cbos setup you would have Meta just saying like well, we're just not gonna spend on ad d And so we're not gonna deliver ad d to fill even if fills the most likely person to buy from ad d We're just gonna keep serving ad a to the next to the next fill or to Andrew right so something like that And by just allowing some ranking it allows meta to sort of extend the reach of better performing ads Instead of forcing each ad to spend on the most likely person who who would buy when when those ads are gonna have sort of an internal ability to reach Sort of an innate ability like almost a genetic component of the ad to reach more people is that it? I'm trying to describe it in a way that's clear I feel like I'm stumbling over it, but is something like that what I'm sound right to you. Yeah, so that essentially each ad set is When you buy medium or meta and you presume that you're gonna use all too big we're asking it to do something and each ad set is trying to do that task to the best of its ability and the best of its ability a portion of that will be The same audience that another ad set is trying to reach Because that'll be the lowest conversion because that that audience is in market They're showing signals frequency is important. No ad set has a frequency of one As a minimum everything is 1.5 or above Yeah, and so that overlap will will happen and there is you know, you mentioned your Your bank media in a different way to other people There was a lot of people out there who yeah, and I'm not saying this is wrong, but there is a lot of people out there who And I'll help them. I'll jump on a call. I'll talk through this Always want to provide some value. There's a lot of people out there that think any activity that they are going to do in the account requires a new ad set Huh every every time every time it's like having a coffee in the morning. It's a new ad set It's a new ad set. It's a new ad. I've got three new ads. That's a launching new ad set and if you do that for six months That's a that's a really problem and then and that's and that's probably because they have this idea that if you add Two an existing ads that you're going to like screw it up right like like people are very worried about this Yeah, I there's a there's a there's a fair from some people from I don't want to talk about some people as they from put them down or whatever But I think some some media buys approach it worrying too much about what could happen in the future Yes, whereas they should think about what's happening right now when we talk about campaign type and they and we sort of about different ideas for setup Sometimes I get the question back of But what happens when I've gotten you creative to test is they well let's try and get this working now and then approach that question in a week's time rather than trying to almost predict what is you know what is coming down the down the truck Yeah, okay First of all, I think you should be more I think you should be more willing to say you think those the people who are saying these things are wrong I think at the point of view is helpful like I I'm having I continue to have This is a broader point for me. I've all tried to get lost in this point too much But there are better and worse ideas about how to do it and the idea that like it just depends on the account And there's no you know, there's no playbook or something is wrong This is the reason we're having this conversation is because like you might what you are saying here about cpm While there is probably some nuance to who it applies to and who it doesn't There's still something objective in the foundation of the way the whole thing works That is important and as either right or wrong or is partially right or partially wrong or whatever And I just I think it's important for people to keep that straight I'll tell you honestly part of the reason that I'm motivated to say this again is that I'm watching our accounts right now I've had just some very fascinating Confirmation of the way that we buy media recently and of course it could probably be better in certain ways And maybe we need to take cpm armors seriously or whatever, but like But that some of the principles we do like I'm watching our some accounts that we just did full media buy overhalls Where all we change is a media buying and it made a real big difference And then I watch other accounts of ours that have left us left our agency And I still have access to the counter whatever and I've seen them change the approach and things have gotten dramatically worse in it And it's why I want to say to people like no, there is actually right and wrong ideas there are better and worse ideas and uh And you should care about them. So anyway, so my encouragement to you is if you think you're if you think you're right You should say it strongly you should tell people they're wrong and you should help them fix it their businesses are suffering because of it So yeah, I'll not know the a couple of bullet points Most people will be better off having glass campaigns in our sets live I agree. I agree. I have bigger budgets and fewer. places. Yes. You chances are you will be better off for that. And this is, this isn't people spending six figures a day. Those are the edge cases or more. And most people will be better off having less things live in the account. And most people will be better off having more of a week, we call it like a weekly OS. So launching creatives on a specific day every week, pausing guards on a specific day, increasing budgets on a specific day and moving away from the emotional reactivity of performance. So if you do those two things, chances are you'll have better performance and you'll make better decisions. You'll have cleaner data. And yeah, you'll be in a better place. Yeah, I think you're right. Okay, so then let's talk about that at the brass tax level. So part of that is baseline. Like don't do this dumb thing of just adding a few assets here and there because you're going to you're going to decrease your reach. Another another possibility, another thing that comes to mind though for me is like, so what do you do when you have a brand with like, I don't know, let's call it 20 products that they're advertising. How do you how do you organize that account structure where you maintain some consolidation? But at the same time, you you recognize that these products have different, I mean, you a lot of times you, you literally, you can't combine products in an ad set because they have two different economic principles of realities, you know, so, you know, I talk about this all the time, but right, like a product with a $200 AOV and a product with a $300 AOV require a really different hack to be successful and even for the same row as and therefore combining them in a highest volume ad set, you're actually going to do real disservice to the out of count. So, so what about that? So how do you how do you maintain some approach to consolidation so that you maximize CPMR, maximize potential reach for the same cost at the same time do that across different products? So you all going to need more campaigns and assets and we have we have accounts where there are more campaigns and assets because there's a business lead to business need to do that, but we aren't then launching more campaigns and assets for sandbox reasons. The approach in the account is is focused on the setup that you talked about the volume of products, the volume of audiences that we can speak to, male or female or kids or pets and we treat it more like a department store and I like to have an idea of having a campaign for each section of a department store or each rail in a shop. You know, you've got jackets, pants, trousers, shirts, having a campaigns for each one of those. Now, you may need to have stronger exclusions, but like proxy, having more products will help you reach more people. So, if you think about it like a fashion brand, menswear, their new product development, they are built on new product development. And so therefore, that is a tick in the box of protecting them from CPMR issues because they're launching products to bring summer autumn winter at minimum, likely even more than that. You know, the whole like drop set up, it's not sometimes it's like monthly or weekly, these brands are doing drops, so that will help you reach more audiences. You will need more campaigns and assets. You may need stronger exclusions and you would then monitor CPMR frequency, new visitor percentage overla. But at that point, I think those accounts are a little bit different because you're focusing on like self-root, which is a very different, people who are selling protein or creatine companies don't really think that much about self-root, whereas yeah. Yeah. Yeah. If you were the kind of operator who is trying to build a seriously profitable, excellently run business, then IntelliGems should be part of your tech stack. Basically, once you're in that sort of mid-7 figure range and growing, you should add IntelliGems and be running constant split tests across your website, across your traffic funnels, to figure out how to make those funnels more valuable for your business and to make that your website more valuable for your business. And the thing is, there's just all kinds of stuff that you can test and optimize to figure out what works. That can be everything from headline, copy, messaging tests, which are super valuable as part of the business. But it can also be larger scale changes. The first purchase offer that you give to new customers, the 10% off that shows up on your pop-up. Should you be doing that or is that just eating away at your margin? What about your free shipping threshold? Somebody asked me on a call today. How do these shipping thresholds hit you? And my answer, I told them what I thought, but my answer over time would be test it with IntelliGems. Go find out which ones are going to work better and see how the shipping charges relative to the order size is actually producing value or not. When you're telling somebody, they're going to have to pay five or ten or fifteen more dollars or whatever, and check out for shipping. That's a real cost, add it on. Is that hurting your business or is it helping your business? I don't know, but you probably don't know either. IntelliGems allows you to test that sort of thing. And critically, to measure it down to the level of profit per visit. And that's so much of why this matters. Maybe you would sell more stuff if you offered free shipping, but it would come at the cost of your profit. Right? Like, I don't know, but IntelliGems can tell you that answer because it ties into your product data, into your cogs, and tells you the outcome of each test based off of the profit per visit, not just the conversion rate, not just the AOV, whatever, the actual profit per visit. So it's really awesome tool. Many, if not most of my clients are using IntelliGems, just one of the things that I think is sort of standard in an e-commerce tech stack at this point. I can go on and on about all the different things you can test and set it up. But the key here is that you should have an ongoing testing cadence and you should be getting the output of that testing at the level of profit. IntelliGems is a tool to help you do it. It's easy to use. You can get 20% off your first three months. If you go to IntelliGems.io and use the code Ferris20, FARIS20, 20% off your first three months with Ferris20 links in the show notes, go check it out. Yeah. So, anything that's just let's also talk about the, okay, so I can, well, so a consolidated approach, I 100% agree with Phil. And I think like for me, this comes down to a much simpler principle, which is like let Meta sort it out. Like the principle is just like give Meta, like as much signal in as few places as possible so that Meta can sort out how to deliver your ad spend. And the reason to separate out campaigns and adsets is because of, is because of economic realities in your business, not because of anything else. It's just that like you have to control inventory differently and you have to control, you know, different cac differently, relative to different products. And so it's about financial considerations, first and foremost. I'm not saying you disagree with that. I'm not saying you wouldn't do it for that reason. I'm sure you do. But it's the main thing there. The CPMR thing is just like, yeah, I'm just sort of going to let Meta reach more people as it sees fit. But also, I'm okay if Meta reaches less people as long as my row ass is still really, really good. And this begins to move us towards the conversation towards sort of the value or not of reach, targeting the value, not of extended reach. And this kind of gets into the disagreement with CTCs of the stuff they published about this, which is like this question of, okay, maybe, maybe. And I think this is the thing I've heard Taylor say a lot of times, which is like, it's actually good that Meta delivers your ad to the person most likely to buy sooner than later that it sort of prioritizes those people. Now, you need to factor in the incrementality of those ads. But, but in fact, you end up generally getting better performance by allowing Meta to do that. All the things being equal, do you actually agree with that or disagree with that concept? I do agree with that in over a finite time period. I do agree with that completely over a finite time period, where it can impact businesses is where the account performed in over over, drawing 2024, the account performed in this way. And in 2025, the account performed in this way. And there was a change. And there was a change in CPMR went up by 10, sorry, CPM went up by 10% frequency went up by 10% those two things combined will have a real impact on CPMR and then your reach will go down. And that's that's really where this becomes more important is when there was a change in the wind. Rather than it's just CPMR being glow consistently. And it's this and it can't you can see the same thing around over metrics where if you've got a $50 CPM, but you've always had a $50 CPM, your business will be set up. Hopefully to support that, you'll have a very high conversion rate, you'll have a very high click through rate, your cost per click could still end up being $1. And you'll get used to that where it's an issue is where your CPM was $10. And now it's $40. Something else in that funnel going from you know spend to conversion. Something else has to afford that increase. Yeah and in most cases it doesn't and where I see businesses really struggle with this and I talked about like the example previous is when businesses are naturally trying to do the right thing. They're increasing a over their increasing LTV, they're increasing conversion rate. And over a long period of time like like two, three years these improvements can be washed out because of rising add costs and that's CPM frequency and CPMR. So if you quite often I've seen brands increase AOV by new product development, but their CPM has gone up and they're just not better off what they're actually doing is staying in the same position, whereas they yeah, where through that idea of increasing AOV to improve their profitability. - Yeah, actually, it's just meant that they can continue to advertise and afford the price increases. So I agree with the idea that high CPMR is in bad where it's important is when it changes over time. - Yeah, so basically in the short term, and I'm gonna repeat this back, you tell me if this is right, in the short term, if I, if I, if I, if my ad is just, if my ads are disproportionately retargeting, my row S actually might be high, and in fact, my incremental row S might be high. So not just a platform attributed to us, but the actual incremental contribution might be high. But if I get stuck in that loop for a long time, and my or, if over time my cost for reaching more people, my CPMR goes up, eventually I'll pay the piper on that. Eventually, what will happen is that I will not have, I will not, like essentially, maybe the way I think about it, the mechanism here would be that those retargeting audiences sort of dry up, and you no longer can sort of squeeze any more juice out of that. And in that case, you actually have now run into a problem that the sort of short term row S wasn't gonna show you, that like you, you know, over the longer term, it creates a problem because over time, that decrease in reach really hurts the business. And now you don't have a pool of people to retarget when the time comes or whatever it is. Is something like that right? - Yeah, exactly correct. And when you think about the wider business, it's really, I find it really straightforward and hopefully a lot of people will do it. It's really easy to see what are these things are important because you think about organic, and follow a growth. You think about your email list growing and your people falling into your welcome flow and you think about word of mouth and you think about direct. And where this becomes really helpful from a planning perspective is at the start of the year, you can look back at your CPMR and any metric for the previous 12 months, what happened versus the previous year? Okay, now what do we want to do this Q4 coming up? These metrics are increasing. How much extra are we going to have to spend just the stay level? - Yeah. - So if we want our Black Friday to be twice as big, interesting, but our CPMR is increasing by 50%. We're going to be to spend more than double because of email. And that's throughout the whole year because it's, and I think the way you're thinking about, thinking about Black Friday is really helpful because you spend all year trying to build this audience and your list and where some brands can get a bit of a sucker punch with CPMR and any of the metric, really like I'm saying it a lot, is you spend all year thinking that you're building this audience and then it comes to Black Friday, you send that email and your email list hasn't grown at the same rate and actually maybe it's gone backwards because you've been getting on subscribers and actually what can sometimes happen is with Black Friday, those returning customers start to make up a bigger percentage year over year. - Yeah. Yeah. I mean, it's interesting, my counter to that a little bit is that when I look at some of the highly, highly seasonal brands that I've run, one of the things I've noticed about them is that like, in fact, I had a meta wrap say to me, like you guys leading into holiday ought to be spending on like some video view campaigns and things like that because you're gonna have, right now you're not gonna reach those people, but you know, spend on that in October, get a bunch more people into your funnel and then November, December, you know, you'll, you know, so it's one of these brands that's like, you know, 40% of the revenue happens in 30 days or whatever, you know, like it's crazy, crazy seasonality stuff. And then I looked at their old metrics and it's funny 'cause this is well before the CPMR conversation that is happening. And I was, I looked and I was like, well, they're actually reaching plenty of people in November, December because the key thing was that when the demand side actually changed because there's more demand for their products at this moment 'cause it's such a good gift. When that happened, well, at the same time, at the same time, as the demand changed, the meta picked up on that and just extended the reach of the ads, you know, and of course, this is interference. So these were all manually bid. I didn't change the manual bid. Meta just quickly determined that, oh, look, all these conversions are coming in under the price that was set at the manual bid. And that means it can extend to the reach of the ads to a broader group. And because of that, like it just extended the reach. So what I said to the Meta rep was like, this problem, this account does not have a reach problem. At this moment, even though they were reaching not that many people throughout the year, Meta was perfectly good at finding those people when the time came to reach them more and it didn't actually have a prospecting problem. Now, so anyway, so I think there's like a weird trade off here where at times I think Meta's actually incredible at doing this, especially with a manual bid, where you set the price and where the demand side changes enough that Meta can sort of interpret in real time. Oh, the conversions are coming faster. I can actually extend now to this person that I would have considered outside the price range. Now it's inside the price range because the conversion rate has gone up enough. And sort of, you know, the bid in my buying setup, the bid cap solves the CPMR problem for you. It's almost the opposite of what people say about these things. Like when people talk about the limitations of cost caps or manual bids, and I don't really run any cost caps, they're only run big caps. But when we do that, Meta is able remarkably to go find more people under that cap as long as the demand side changes or as long as the creative is performing especially well or whatever it is. You know? And so I just have just not seen a real reach problem from doing those things. I don't know if you have any thoughts on that. It's sort of a specific use case. But yeah, I wouldn't be trying to artificially improve CPMR before-- Yeah, interesting. Before hand, and actually the before you mentioned the Meta wrap-out, I was asking if I need to make a point about this, but running non-conversion campaigns is the last thing. It's not the last thing we do. Once we've done everything, Kels, and ideally everyone else is in a perfect order, we don't ask the need to. At that point, we do have tactics and an approach to doing it, but it's the last thing we do. And in your example where the manual bid isn't spending and then is spending all of the sudden, it's the right reason. We would be using CPMR from a diagnostics point of view at the end of that period to understand year over year was the performance better or worse. It was 10% better. Why was that? Because the product was better. Was that because the creative was better? Or did we actually find an efficiency in the cost to advertise? And that afforded us a luxury that we didn't have the previous year. Or the other way around, it was 10% down. OK, why was that? Was that because the creative was worse? Or because the account did more retargeting-- so it's more of a diagnostics thing to help you make the right decision. And the idea of saturation and time, I think sometimes people can struggle to quantify what those things are. Frequency and CPMR reach can put some black and white noise in front of you. All right, so let's give some people some really simple practical stuff and then let's also give them some bigger picture stuff. So let me rattle a couple ideas off that I hear from this conversation from your thinking about this. That I can help. So the first is go look at your reach first. Just like go check. Are you reaching? Is your reach changing relative to your total? I mean, really, the question is almost reach aside from cost because it's possible that you're costing more money to reach more people, but you're still reaching more people on its worth at in the end. So go look at your reach. Do you think that's a good start? Just go check and see if you see any patterns in your reach. That's a practical start for this. Yeah. And the two time frames I would look at, because I think the time frames are really important, is year to date versus the same time last year. Yeah. And then you compare that with spend and reach is reach flat, but we've spent more that that can tell you something. So year to date versus the previous year. And then also last seven days versus the same time last year. Seven days is a very small time period. But and you could increase it to last 30 days, but a smaller time period versus the same time last year, especially if there's an element of seasonality, which that is for most bonds. Yeah. Yeah. OK, if you see a problem there, I see two solutions right away. So if you see yourself reaching less people. Now one thing I do actually before I see more practical, one other thing, do you think it's possible also that meta's efficiency is getting better enough from in terms of ad ranking and ad creative delivery that one reason for less reach is that meta is getting better at putting your ad in front of the right person faster with less waste. Potentially. Yeah. I mean, the platform is really good, especially even from like a user perspective. The relevancy of the ads in my feed is very strong. Yeah. So I think the theory is right now is that meta has to do this because they already got all the users. Right. So like if they want to grow their revenue, they have to make it so that each user is worth more money to that, which means they have to make the efficiency of your reach better, which would actually push your reach down without affecting your actual impact because the more meta can match the right ad to the right customer, the more they can charge for each impression. And they, like I said, they already got all the users. They can't get more to you. Everybody uses it. Who's going to use it? So, you know, I don't know. Maybe they can make threads really, really big. But like otherwise, this is just top. So. Okay, so there's that. Okay, let's say you see a problem though, with all those things. I see two things that are first steps. Tell me if there's another first step, tell me to disagree with these. The first is consolidate, right? Less campaigns, less assets, and I'm gonna ask a specific question here, should people get rid of their creative testing campaigns as part of this because that would be campaign bloat or not? It depends what else, it depends what other campaigns you have as well as creative testing. I think creative testing has definitely has a place in a lot of accounts, and especially if you've got a lot of really good creative, then it just depends what else you've got. I would also depends where you'll spend this. Okay. Okay, but generally speaking, more consolidation. Okay. And then, let's talk about this from the creative perspective. The other thing that you might consider doing is check your creative, is it stale? Do you need a higher volume of high quality, highly diverse creative, and you can add anything else that you want to that. Anything else besides consolidation and creative diversity, volume, and quality? Exclusions. Oh yes, okay, yeah, good. Yeah, more about that. Yeah, do we need to increase exclusions? I feel like when I first thought you'd get into media buying a decent level, you know, 2018, 2019, listen to the great Andrew Faras podcast. (laughing) I feel like we used to use the all of the exclusions possible. We used to use social media, engage your exclusions, page follower exclusions. We used to use all of them website visitors, obviously. And then post iOS 14, it sort of scaled back to purchase and email list. And I think that's where most people are. And in most cases, that should be enough. You may need to increase, you may need to include website visitors is usually the next protocol. So exclusions, maybe need to save them. Right, those people and put them in a different remarketing campaign. Is that what we were saying? In some accounts, we'd in a small number of accounts, we would do that if you need to do that. If the frequency will never be one, if you're using NorthBee more triple well or at any of these tools, you can see new visitors to percentage and it'll never be 100%. So we always do some level of retargeting. So exclusions. And then creative, one thing we haven't talked about is. Sorry Phil, just for exclusions, just to be clear about it. So what are you saying? Paul, those retargeting audiences into a separate campaign specifically so that your true prospecting audiences don't get stuck in a loop on them. That is one thing that we're testing at the moment is, and we haven't spoken about this. I've got a couple of tweets about this online. And we're testing it in a maybe like five or six accounts where we have a retargeting Godset, where the purpose of that adset is as a distraction so that meta doesn't serve those audiences in top of funnel. - Yep. - Because if we're not serving that those audiences in that adset, meta will try and serve them elsewhere. Yeah, we're collecting data on like that approach, but we have seen it perform really well at so far. So yeah, I can- - What are we gonna say about creative? Placements and platforms. - Okay, yeah. One of the first things that we used to do a lot more after when we were auditing accounts is, okay, so where are you spending with creative? Is it Facebook feed? Is it desktop? It's all one by one. - Yeah. - Okay, well now stories and reals, reals, obviously like the success story, maybe like that was three years. That's where it's growing by 16. So look at where you are spending. Look at your frequency and CPMR per placement and per platform. There's a potential that you'll find a real interesting data point there that maybe your CPMR on Instagram reels is tiny. And you could massively increase reach by creating more ads that serve to reels. I'm not saying go out and make the natural media buying approach there is let's make an ad set with custom placements and for spend into reels. I'm not saying that. One thing that we'll do is we'll only upload videos in nine by 16 tall so that we don't give Facebook a, or meta, sorry, a four by five feed asset so that method hopefully defaults to reels or stories. - Okay, any other super practical simple things like that? I like that point about placement diversity. I don't know if it's good to get that. - Partnership ads. - Oh, okay, good. Yeah, yeah. Yeah, one of the big things that we've seen for a long time is partnership ads can get a cheaper CPM. I don't want to start an argument about, you know, what CPM doesn't matter and low CPM. - Yeah, sure, sure. - But if you've got really high quality seat partnership ads with really high quality creative and a large volume of partners posting regularly, you may get a benefit in CPM and then that can now both help from a marketing experience point of view you're reaching, you're taking advantage of their audiences but you're also getting a benefit from meta and for a long time we did used to see meta. It felt like meta favored our activity because we were using partnership ads and you know, maybe it's a little bit of a theory but meta used to be very focused on shopping. - Yeah, shopping feet, I believe they moved to partnership ads and partnership content instead of shopping. - Yeah, interesting. Okay, and then let me give you one scenario here and we'll wrap up, okay, so assuming those simple things are done, simple as quote unquote, there's still hard to develop, good real ads and things like that. But okay, so let's say I've got a brand and this is an actual brand that is finding extremely high frequencies and low reach. They have conceptualized their customers as existing in sort of three concentric circles, okay? Inner circle, super core customers, second circle, circle two, people who are relevant to them for sure but maybe a little less super core and then circle three would be the people who they eventually would like to reach so they recognize they probably can't reach them that effectively right now but they're tangentially related to the community they're in, et cetera, okay? So you get the idea, right? Three circles, they're there, they have done all the things you've said, right? They're generating a high volume of high quality, diverse creative, all that stuff. They are consolidated, they are running some partnership ads you know, they can always do more of them better of course but there's no glaring issues with any of those things in the account, okay? When I do this and when I go on my coaching call with them in about two minutes, okay? What just happened? What I have said to them is you don't have a media buying problem, you don't have a creative problem. What you have is a product and marketing problem and you have to now think about this at a much broader and harder level and we talked about this a little bit earlier in this episode but is that the reality for a lot of these brands running into this fill that actually at the end of the day the challenge is something much larger which is that at the end of the day there really is a customer and there really is a product and you have to be able to reach different customers than you're reaching right now with your product if you want to expand your reach. I would be asking them to talk me through what are you doing outside of the meta-out account? Yeah, I'm ensuring that that I just stays level that consistently doing these things outside of the other account or ideally they're doing more of those things same as how you could launch 10 creatives this week okay, that's launched 20 creatives next week do the same thing but outside of the other account what else are you doing to help meta and then measure the impact of those things inside meta? Yeah, I like it. I think that's really good. That includes product development, right? Yeah, product development, seeding, giving products to people to post about your product and tracking when they post is quite worth in measuring the impact on meta. blog, SEO, GEO. Yeah, yeah, sure. All of these things. Yeah. Phil Keel, as I said, is the director of paid social at Tycoon Digital Phil, that means helps brands, e-commerce brands, scalar businesses. If you want to work with them, I'm so happy to recommend taking a call with Phil if you've got the CPMR problem. He's looking at CPMR problem out of council all the time and could probably help you with it. You can go to TycoonDigital.com or even better, Phil says, just reach out to him on X. Go check that X feed. There's a bunch of stuff there about this topic where he's really broken it down in a lot of details. Shoot him a DM. DMs are open, Phil. Yeah. Yeah. Great. Shoot him a DM. He'll check that out and see if he can be some help to you. So go follow him. Go see if he can be some help. Phil, delighted to have you and delighted to have this conversation. Thanks for taking the time. Thanks, I'll do. [MUSIC PLAYING] If you want to follow up with Phil and see more of how he's thinking about this, he's published a lot about this on X recently. You can see his interaction with Taylor Holiday and CTC about their study that actually took kind of a counter position to fill. All that stuff is really easy to find on Phil's X feed. So go give him a follow. Go start scrolling, find that. And yeah, the link to that's in the show notes. Of course, I'd love for you to subscribe. Like I said, if you liked this episode, you will like a lot of my content that is full of media buying stuff like this. We're really trying to understand the best way to do it. I think media buying makes a huge difference for businesses. I care about these conversations. I want to do these things right because I care about doing a great job for brands, trying to grow them profitably and quickly. And stuff like this really helps you get it that. So I'm recording this stuff all the time. Definitely subscribe wherever you're watching or listening to do that. You can also email me podcastedagfgrowth.com with any questions or thoughts you have. I'd love any suggestions for episodes, like if you've been Q and A episode or something like that. That would be really fun. Go mail back episode. So email those to me, [email protected]. You can also go to agfgrowth.com and still let the intake form there. Just tell me about your brand. If you want us to work with you, I'd love to hear a little about your brand. Just tell me a little bit about it. And I'll get back to you and we'll see if there's a fit for us to work together, grow your business together. We'd love to do it. Don't forget to subscribe also because I have a great episode coming up with Ezra Firestone very soon, which is always great to have Ezra on the show. It's such a legend. It's going to be great. I think that's going to be out next week. So thanks so much for watching, for listening. Don't forget to follow up with my friends at more staffing and at Intelligen's. Go check both those out. Thanks so much. Talk to you soon.