Pixel Perfect: Is Your Tracking Lying To You?
This episode dives deep into how much your tracking setup actually controls your Meta performance, and why most brands only pay attention to the pixel when things break.
Brett walks through his journey from old-school media buying into early Facebook ads and how that evolved into Tag Hero, a specialist shop focused purely on technical conversion tracking across Meta, Google, TikTok, Snapchat, Pinterest, etc. He explains why data quality is the real throttle on performance and how you can have the best creative in the account, but if Meta isn’t getting clean signals, the algorithm just can’t do its job.
Key Takeaways:
How you can tell if your Shopify → Meta tracking setup is actually “good enough”… or silently leaking conversions.
If Meta says you’re getting 8–13% more events with CAPI, are you actually set up to capture that lift—or just assuming you are.
The easiest way to verify your pixel and CAPI events are deduplicating correctly instead of double-counting purchases.
The Event Match Quality scores you should care about, and when a score is good vs a legit red flag
Why this one tracking perameter is arguably more valuable than email for optimization.
Why optimizing around a specific EMQ score might just be gaming a number that doesn’t move revenue.
When Shopify’s default Facebook Sales Channel App is the right call and when it might be time to graduate to Elevar/Blotout/EdgeTag-style stacks.
The incremental tracking upgrades that actually pay for themselves after you start spending 6 figures or more a month in ads.
The truth of what exactly happens when Meta flags you as ‘Health & Wellness’ brand and depricates your lower funnel.
How you can still win in health & wellness categories on Meta
The key steps to confidently rule out tracking as the culprit when your ads stop working so you’re not re-building creative for a data problem?
This episode is sponsored by Northbeam, the marketing attribution platform that we love here at Scalability School. If you’re ready to cut through the noise, stop guessing, and actually see which ads are driving your business, book a demo at northbeam.io/demo, and tell them Scalability sent you. Join the club.
To connect with Brett Fish over at TagHero, you can DM him at https://x.com/brett_fish or head over to their site: https://taghero.io/
To connect with Andrew Foxwell send an email Andrew@foxwelldigital.com
To connect with Brad Ploch send him a DM.
To connect with Zach Stuck send him a DM
Learn More about the Foxwell Founders Community head here to learn more.
Full Transcript
It's a quantitative way for meta to kind of tell you how well, how confidently can we match this event, this action on your website to a user on our platform. That's, that's the gist of it.
Um, so, you know, it's, it's what data are we passing back to help meta make those matches, which drives optimization, which obviously helps your ads performance and efficiency. One client referred to it as a mysterious force, which is, I think is about as specific as we can be. Meta has not produced any like documentation or comms on this, but you know, once we saw it across. Dozens and dozens of accounts, you're like, okay, this isn't just a one-off. This is a, this is a trend.
This is a thing where spend dropping conversion rate, dropping CPA, like going out, like negative impact on the ad account, maybe introducing some upper funnel events just will sort of tweak and shock the system a little bit and go after a different pool and sort of cohort of users on the platform and bring them in. So I think like optimizing for that stuff is definitely worth a test, like optimizing for upper funnel events.
So I think in that same light, if you want to optimize for a quiz completion, that's pretty high intent. And now let's take a listen to the scalability school podcast. Welcome to another episode of the scalability school podcast. Um, pixel perfect with Brett Fish from tag hero Brett. So glad to have you on the show.
Honestly, friend of the pod, um, friend in life, actually, Brett's been on, uh, one of my family vacations this year already with my, with me and got another one coming to December. So, you know, that's, you know, we, we have really gone full circle here. Um, but we're excited to have you on you and Zach. We're talking about a whole bunch of stuff with pixel and Zach, you want to give the backstory about why you wanted to have Brett on? Yeah.
I mean, uh, TLDR is we were concerned, um, about some just like auditing our pixel, essentially we were starting to have some issues with performance and we were trying to back into, if it could, if it could be any errors that we were firing on, on the pixel. And then we, uh, we hired Brett, Brett and his company and they came in and. Did a bunch of cool shit and I didn't understand it.
So it was like, Hey, Brett, can we, uh, have a conversation so you can explain this to me like I'm five. And that conversation was incredible. And I was like, Hey, you have to come on the podcast and explain all the cool things, Cappy, EMQ, all these acronyms that get thrown out there that I didn't know what they were, uh, but wanted to, wanted to know more about them. So that's what the episode is going to be about today.
I was trying to think about like what the, what the, what it compares this to in real life, because I feel like a pixel is not, you know, the pixel and pixel setups is like, you don't realize outside of the basics, really, you know, it's important to have a pixel set up, but you don't realize how, how bad it is until it breaks.
It's kind of like insurance. That's why I came up with, I don't know if it's a great comparison, but it's like. Shit, it broke. I don't have this thing and I don't know what to do to fix it. Um, and, uh, yeah, no, I'm excited to dig into this. So Brett's gonna, Brett's gonna tell us why it's important, uh, how to fix it before it breaks.
This episode is brought to you by our friends over at Northbeam, the marketing attribution platform that a lot of us at listen to scalability school and the host of scalability school, you know, we swear by and use Northbeam rolled out something brand new that I think is pretty awesome. It's a real unlock, I think called clicks plus deterministic views. And it's the first ever deterministic view through attribution model.
I mean, let's be honest, right? One day click had us flying a little bit blind, you know, the drill, right? Your Tik TOK or your CTV campaigns are building awareness, clearly driving site traffic and lifting sales across channels, but your attribution dashboard still gives all the credit to Google search. Cause it was last click. It's like tracking who made the final shot in the basketball game, but ignoring the five assists, right? That got the ball there. So clicks plus deterministic views flips that it ties actual impressions, not
guesses, but verified impressions to real conversions. We're talking direct integrations with meta, Tik TOK, snap, Pinterest, and even CTV, and Northbeam uses hard identifiers like order IDs and hashed emails to prove that someone saw the ad didn't click, but later bought. That means you can finally measure the real impact of upper funnel campaigns.
And the results are pretty nuts. One brand saw 283% more attributed conversions on Tik TOK. Another got 175% more attributed revenue than they knew even existed. And you know, another one drove 59% row as lift on Pinterest. So if you're tired of under budgeting awareness, because attribution can't keep up, this is your chance to see the real journey in the full journey.
So Northbeam click plus deterministic views is live now head over to Northbeam.io forward slash demo to book a walkthrough and tell them we at scalability school. So, I mean, first of all, Brett, anybody you want to flame, uh, and talk about on the internet, I'm just kidding.
Who do you hate? Who do you, who, who do we hate this week? Um, who's driving the spunkers? I don't think I can lead in with that. You might have to warm up warm up. That's just, that's just, uh, that's just to get everybody laughing. We're, we're good. We're, we're all friendly people here. Uh, it's all love. Um, Brett, uh, do you want to tell everybody about like how you got into this first, like briefly? Like, how did you start doing tracking so that, and like, what is it that you do now? Exactly. Yeah, absolutely. Uh, well first, uh, pleasure and a privilege to be on with you guys.
Um, highly respect all three of you on the work side and then getting to know you all, you know, in real life as people too, it's been awesome. So it's a privilege to join you. Um, yeah, so it's, it's totally accidentally is how I'm like here with Tag Hero, uh, my background was all in traditional marketing. So like, I'm, I date myself.
I'm old, like my, my original work in marketing was like old school media buying print, radio, TV, billboards, all that stuff. And then my career are kind of coincided with the emergence of digital. So I started buying digital banner ads and stuff like that and saw the tracking and the potential of like, you know, a billboard, you put up a phone number and you hope someone calls it with the banner ads, we could see clicks.
And every, I'm like, this is amazing. Had the opportunity with some early job roles to spend money and work heavily with early Facebook ads when they were right column classified ads, power editor era and days. Um, and so I got to experiment with a lot of work on it.
And after I'd done that career stops a couple of times, I like, I know what I'm doing here, I'm pretty good at this and, and split off to do my own thing, start my own paid social agency. I was like eight or nine years ago. Um, and just a couple of years into the agency though, we were constantly having issues with the stuff we're about to talk about, like technical tracking, pixels, backend stuff, and you know, now what, like seven years ago, six years ago, previous era in digital marketing, right? Even less available in resources than there are now. And so I got sick of it.
And so I'm like, okay, we're figuring this out. This is not going to be a pain point anymore. Cause it was hindering our ability to serve our clients. And that was the main thing. And, um, did some networking, brought some people in from outside that have like dev backgrounds and stuff like that and kind of built it into a core competency of our agency and then very sort of, uh, timing wise coincidentally, right around the same time we got pulled into a program with meta called their technical partners program. So we were a third party, um, trained vetted and paid by meta to help their
advertisers with all the tracking stuff on the backend. Cappy really wasn't a thing back then at that time, but pixel catalog feeds, all that kind of good stuff. So we did that for about four or five years. And when meta kind of quickly and unceremoniously ended that program, we had a lot of people coming to us like, do I have to run ads with you to have you fix my tracking? And it was like, it was so awkward. It was like a sub bullet point of our agency. And that's when I decided to spin it off and do its own thing as tag hero.
And that was almost three years ago. Uh, so, and then, you know, it's taken off since then and there's clearly been a need for it. So now I'm fully committed and a hundred percent of my time is on the tag hero side, helping people. So, um, when we talk about all this stuff, like pixels, server side event tracking, all the backend conversion tracking for platforms like meta and Google and Tik TOK and Snapchat and Pinterest and all the others, heavy Google tag manager stuff. Um, you know, all the backend technical conversion tracking stuff that drives
these platforms, that's our expertise and what we can help, uh, brands and advertisers and agencies with. Yeah. And you're the technical partner, chief technical partner of the Foxville founders membership. And they've helped literally every member probably with an issue because they don't stop these fricking issues. Um, so constant.
Yeah. It's like, yeah, Brad made a great point. Like no one cares about tracking really until it breaks. It's one of those kinds of like given things that you expect to work until it breaks, but when it breaks, what you realize, and, and I've always been adamant about this and why I went full in on tiger was it's such a foundational part of success on these platforms that has to be in place. You can produce the most banger creative that's ever been produced.
And if the platform's not getting the right signal off of that, it's not doing it justice. Um, so creative and CRO and offers and all super important to success. But at the end of the day, these are all algorithms, right? Good data in good output, garbage in garbage out. So the data and the signals are really critical to success.
Um, but people don't really care until it breaks and something goes sideways. And then that's when we usually come in. We'll talk a little bit later about how we're trying to get ahead of that. Brad talked about insurance, but, um, yeah, we, we help them sort that stuff out when it's not working like they expect.
Yep. So let's go ahead and just jump right in, right? Like out of the gates, actually, Brad, you had a story you wanted to talk about. Yeah, I have two things.
Um, Brett mentioned things breaking in literally the last seven days on meta has been like literally the entire platform just keeps having perpetual disruptions, I think ad spend for all of our clients stop for like two hours or like an hour, at least in the middle of the day today. So fun, fun timing on things breaking completely unrelated to the pixel or the pixel things out of our control. Um, I just want to share something I saw this weekend, uh, or actually heard from a friend is like somebody outside the industry.
And so that makes it even more compelling. My ears start to perk up when I hear people like interacting with brands in real life, because I spend all time, all my time talking to, uh, you, you two, you two morons in particular, volunteering, of course. And so I was like, okay, you're perpetually in the weeds.
You don't really, you don't really take a step back and zoom out and like, see how people interact with stuff. Long story short, sitting next to my friend that I've known for since, since, um, elementary school at a wedding and he starts telling me about his sunglasses and whatever it's like a normal conversation about some glasses. He's like, yeah, you know, I just lose them all the time.
So, but I'm really, I have a, I have a pair from shady raise and I really like them, um, and I have their, I'm in their replacement program. So I just like, they just replaced them for me for free. It's like they're $75 sunglasses and they replace them for free. And I was like, Oh, that's cool. Like completely for free. And he's like, well, no, I pay shipping.
And I was like, okay, how much, how much do you pay shipping? He's like, Oh, like 15 bucks. But like they're $75 on glass. They're like, okay, these, these, these sunglasses probably cost them $2 to make. They probably cost like $7 in shipping. And most of that's like, because they have to go into a box, they don't get crushed in the mail because they weigh nothing.
Um, they're making margin on the replacement program and they're, but like the way that he worded it to me was he was so gassed up. He's like, I get my sunglasses replaced for free. I don't have to remember it. I don't feel bad about this. It's like his perception of their replacement program was I get free sunglasses.
My perception is they're making money. It's like everybody loses their sunglasses. It's incredible. I like, I like that the first time you left your house since we've started this podcast, you have a story. Yeah. Oh yeah. Look, I'm going to Vegas this weekend for another wedding.
So who else might not interrogate about their D to C shopping experiences? So yeah, I thought that was really fascinating. That's that's fascinating. It's good marketing. So, uh, Brett, like let's just jump in and talk about, um, what does the essential setup look like Shopify default? Does it cover meta? Okay.
Or does it cover like what, what are the issues with just a default setup? Yeah. Um, no issues with the default setup is I think the main thing. Um, I remember a while back there was kind of this big wave of concerns and questions about Shopify's app and the tracking and to be fair, the apps, you know, it's had its evolution and changes it was originally owned by Shopify. I think now it's owned by meta or vice versa.
Like it actually changed hands of like who is owning and maintaining it. But, but the long story short is. Look, if, if you're the majority, I'd probably guess of Shopify brands and stores out there and you know, you do decent business, but nothing crazy. I think the app works great. You get pixel and conversion API server side tracking out of it.
It passes back all the parameters to match up with your catalogs and all that stuff. So from like a data standpoint, it's pretty solid. Um, does it squeeze every last single drop out of a potential data connection? No, but, but, but it's a free, very quick and easy to set up app.
So I think for a lot, we, you know, we, we always recommend to go that route wherever possible and wherever it makes sense. Um, you know, one of the things with it is it's designed for out of the box, right? So if you want to do anything custom outside of the typical e-commerce funnel and flow of page view, view content, add to cart purchase, all that stuff. If you want to do something custom, it's not going to do that. It's kind of just intended for out of the box e-commerce, but it does that well.
So is it perfect? No, it's not perfect, but I don't think you're, you know, at a major disadvantage or it's detrimental if you're using it. I think it works just fine for a lot of brands and takes, takes a lot of the guesswork out of it, because anytime you go outside of something, you start doing custom things, there's more things that can to be considered.
But, uh, no, I think it's pretty solid. Um, and for a lot of stores, it's, it's, it's totally adequate. If you're at a size and a scale with traffic and volume and revenue and everything, where a small incremental lift in your overall kind of data would offset the cost of putting that all into maintaining it, sure. Explore.
There are other solutions that are a little more advanced, but like I'd say for the average store and advertiser, it does a really good job. I want to come back to what you're saying for like, when you want to squeeze things a little bit more and like when that starts to make sense. Uh, but coming back up one step, like, okay, so the Shopify setup, I think most people listening to this will understand, but just for the sake of making sure we, we, we get everybody that's literally, you go into Shopify, you go into sales channels, you install meta and you just like go through the
setup process, you select max data sharing or whatever your settings need to be and that, that sets up the pixel on the front end, um, saying boards that might not fully understand front end and then back end is conversion API. Um, is that fair? Correct. Yeah. It's tight.
It's called the Facebook and Instagram Shopify sales channel app still, but yeah, you go in and in the settings, there's a section for data sharing and you can connect your business manager and your dataset or pixel ID and everything. And, um, you can, I recommend setting it to maximum. There's a couple levels set it to maximum data sharing. And that'll handle, um, to your point pixel, it'll handle server side events.
And one of the things that we can dive into layer, but one of the really important things when you have pixel and server side is that they deduplicate properly because I didn't an ideal setup from meta standpoint. And we were talking about like, what is an ideal setup, Sam set up look like.
You're supposed to send a pixel event, which is a front side event and a server side event for every event. So if I go to Zach store and I go through and I make a purchase that purchase event for me should send a pixel and a server side event, but they should both be carrying a unique identifier called an event ID to deduplicate. So it doesn't count as two purchases. Cause I only made one purchase. That's the best practice set up to ensure you're capturing that event.
Um, but if you don't deduplicate them, then you, then you have over counting and over reporting and all sorts of issues. So, um, the app handles all of that for you, which is why I think it's, it's a really good option. So it'll automatically take care of all that, but correct. Yeah, Brad, you, you set it up, connect it and you're off to the races. You really don't.
Um, actually, Andrew, there was a founder's member the other day who was like rebooting their store and we went through and set it up and like, I think it was like a minute and a half on a call. And he's just like, Oh, like that was it. I was like, that's it buddy. You're all set. So, um, it's really straightforward overall, but yeah, it works well. Nice. Cool. So that, that covers it for the vast majority of brands, but there's obviously people, um, like Zach who want to crank it up a couple of notches.
So like, what is there like a threshold for when it starts to make it like, okay. So I'm sure there's a couple of ways to look at it, like thresholds and revenue traffic or however you want to define it.
And then there's those like unique scenarios where you're like health and wellness, which I'm sure we'll come back to, or you're doing something custom, like maybe some like quiz stuff or, you know, other stuff that we could, I'm sure we can, we can talk about that too. But, um, any way that you like think about, okay, we need to maybe go a little bit deeper than this, like, how are you, how are you looking at it? And how are you advising folks on when to do it? Yeah.
But, you know, there's, there's not really a line and actually Zach would probably be more equipped to speak to that than I would be, but there's not really a line as I see it, it's just, if you get to a point where you're like, you know, we're at a certain level of size and scale, we want to make sure our data is, you know, cause some of these tools that aren't the app, you know, the app is designed to do exactly what it does.
Um, some of these solutions out there, um, L of R blot out edge tag, all like there's other things because that's not pure tracking. I mean, it includes tracking, but there's other things that software can do. It can do like user fingerprinting and session stitching and some other things. And, and what I always tell people is. Will you get more purchase events? Not necessarily.
That's not what it is. It's like you're capturing more purchases, but like you might have more user data on those upper funnel events that tells the platform, you know, for prospecting campaigns and things like that, who's hitting your website more effectively and helps it optimize.
Like that's what I mean when I say incremental, it's not, you're not necessarily going to capture and report more revenue, more purchases and things like that, those events should still fire and be captured in a more regular setup. But those, those tools can add additional things from a SaaS standpoint. They can enrich your data and squeeze a little more out of it. Um, I always consider that somewhat incremental.
So it really comes down to me to what is the cost of that platform cost? Um, with some of them, what's the ongoing like upkeep? Like we, I can tell you, we audit a lot of tracking of. We signed up for this tool and we don't use it anymore. And we're not sure what the aftermath looks like. Like we get a lot of that. So some of them can be technical and require some upkeep. That's more than some brands are equipped or willing to do.
So you have to kind of consider all that. But if, if, if that all makes sense and lines up for what it gives you, then, then I think it's could be worth it to try. Um, but I don't have like a hard, like at this dollar of revenue or at this amount of traffic or events, does it make sense? There's no like, like hard line on that.
It's just kind of, I think when it starts to make sense for you as a brand to look at it. Yeah. I mean, the way that I justified is like, if you're spending a few hundred thousand dollars a month on ads, probably should take a look at this. Right. And make sure everything's dialed in. Right. I think that that's just like the baseline thought on my end. Oh, yeah. I mean, I think let's, let's get into like some of the other things that you can do.
Right. So I remember like one of the, one of the questions that we had for you, Brett was, is there a way for us to see and run pixel events for a new customer purchase versus returning customer purchase? Cause maybe we want to test optimizing off of that to see if we can get focus on new customers, right? Is there, is there some level of event data or something that's happening on the backend that might be able to actually show better performance if we just opt to optimize off of new purchase? Like tell us more about the other things that we can start to do with like
going a step deeper than just the standard setup. Just to at the high level. Yes. You can definitely capture events, um, either through like a platform, like an L a bar or a blot out Shopify actually has that as part of their core web API.
So even if we do something with like Google tag manager, we can still determine was this event from a new customer that's never bought from you before, or is it not a new customer? Right now you're returning customer and you can fire like a custom meta event. For example, um, you could even do a custom Google conversion if you wanted, but, um, you know, you can fire those respectively where you say, this is a new customer purchase versus this is a returning customer purchase. And then you could choose to optimize for a new customer.
That's the technical side. My opinion and soapbox on it is I don't actually know how useful that is. I'm always at like a results talk, right? So if it works and it works and we can repeat that like any other kind of experiment, like great, like what does my opinion matter it works, right? But my thing with that has always been there's no metric in metasys.
And I think we talked a little bit about this Jack, Zach. And you mentioned something that you had talked about on a call as like alternate reasoning that I think would be worth like going through, but from the technical side, there isn't a metric in meta system that says it's new customer that you like. There's no recognition and metaside like, Oh, this is a new customer.
Um, so when you're passing a new customer event, you're really just passing a subset of your overall purchase event. And is meta going to be able to optimize toward that group of users as a new customer and go find other people who have not bought from you before? Not necessarily. So I know some people that have tried optimizing for it and it worked for a little stretch and then maybe kind of, you know, petered off over time.
Um, I've never been super sold on the technical backend of it, but it's definitely possible and so we try and it's worth testing. For sure. We just wanted to try it. Right. Similar thought. It was like, let's just see what happens. No bump in performance. So like, I don't think, and again, that's just one dataset to use is just us, but same, but for us, by the way.
Okay. Got it. Yeah. I mean, this last week, we tried it. Didn't do anything. I think optimize enough of the main standard purchase event is probably the way to go 99% of the time. Right. So, um, yeah, I'm glad that you called that out though. Yeah, that and exclusions, right? I mean, using as robust of exclusions as you can in your ads, I think is probably a better overall approach than just trying to optimize toward an audience of people who it's their first time buying from you. Um, and, and I could be wrong about that, right? Meta could have something in their infinite data point structure that says
like this person interacted with this store for the first time or whatever, but we haven't really seen any proof or evidence of that. Um, so it's all kind of speculation. Honestly, I think it's been a little bit more almost marketing fodder for like data tools to say, Oh, we can do this. Then it is like real actual practical execution and like results off of that.
But, um, it's like, I've heard some for some periods it's worked. So it's definitely worth testing and trying if you want to. And it's definitely something that's available, especially if you're on like Shopify, um, this is a, this is an ask for a house to sponsor the next, uh, I don't know, handful of podcasts and give me house for free so I can technically trial that, you know, trial this out and run it by, uh, you know, the book.
But outside of that, yeah, we tried it ourselves. No difference. Um, should we get into EMQ? Um, yeah. So the one I would love to talk about EMQ and just like really briefly, can I actually let's talk about Cappy first. If you don't mind, um, you have the default install and just so I understand what Cappy does.
Cappy goes on your site and if someone is filling out a form or puts in their email to capture it, it tags that together. Is that what Cappy is part of what Cappy is? What is Cappy? No, it was a good try. It was a good try Foxwell, but, um, no, so, so what Cappy is, is it's an alternate route to send your data to meta. Then the pixel, the pixel we all know and love that's been around forever.
So it's an alternate route. Um, the pixel is JavaScript. So it's code that lives on your website and it shoots data directly back to meta from there. The issue with that is it can be disrupted and blocked by all sorts of things. So like browser load errors, slow internet connections, ad blockers, JavaScript blockers, there's a myriad of things that can disrupt that pixel sending that signal back. So Cappy stands for conversion API. And it's, it's met as terminology for server side tracking, but it's essentially
a, instead of sending it via JavaScript via the pixel to meta, we're actually going to capture that on our own servers. And then our servers are going to connect to meta servers directly with an API and pass them that data. Right. And it's capturing, it's capturing the things that they're doing though on the site. The same events.
It's all the same data and all of the same events that the pixel sends. Correct. So that's what I meant. Like it's capturing action, not related to the pixel. I didn't say that properly, but it's like, if they put in their email or they start filling out a form or they click add to cart, like it's capturing all that information just independently of the pixel. Yeah.
It's same information, but sending it to meta different that's less likely to get disrupted. Okay. I was thinking about if I was going to ask this on the podcast, because it risking looking stupid, but I don't fucking care. So I'm stupid. So I just wanted to make sure like that's what I thought. And I want, so that's what Cappy is. Is there anything that people can do to help improve the Cappy experience? Uh, I think one main sort of distinction to make up front is, is that Cappy is like server side events when it comes to emails and stuff like that, that's EMQ.
And that's what Zach mentioned that we can definitely dive into in more depth in a minute. Um, when it comes to Cappy, really, you just want to make sure you have it implemented. So the numbers kind of move, but metas official documentation has always been you get somewhere between eight to 13% more events by having Cappy in place alongside your pixel in a redundant setup than just pixel alone.
So Cappy is really valuable in that sense of you would have some signal loss if you just were using the pixel. Whereas if you have Cappy, I mean, let's say ballpark nine or 10%, that can be a huge swing as far as like revenue and purchase of activity reported and things like that. Um, in regard to make making it like, I guess, as effective as possible, you just want to make sure it's a, it's a strong, robust Cappy setup. It's a true redundant setup.
So like the Shopify app would take care of that for you. If you're using Google tag manager or something like that, just making sure you have a Cappy implementation in place alongside your pixel that captures all the same events, you should be in good place. It's it copies really a backstop to get any of those events that the pixel loses or drops or their signal disruption.
How would someone know? How would they know if the pixel, if like the Cappy setup isn't good? So, so let's say I have the default installed and I'm like, I feel like I'm not doing good here and then I'm like, see an ad for, you know, one of these tracking things I'm like, should I sign up with it flat out or whatever? Like, can I just, where do I look to solve this on my own? Yep. So the main, the main place you'd want to go where we spend most of our time is in the meta events manager.
So if you go into meta business manager, um, the events manager tool in section will show you all the events that are coming into your site. And in that overview on that dashboard, there's a column that says connection method, and if it says multiple, that means you're getting events via both pixel and Cappy, and you can actually expand any of the individual events. And there's a chart, there's a graph that's actually really useful.
And it shows you, here's the server events that are coming in. Here's the pixel events that are coming in and they should be pretty tightly correlated. So if that all is there and is in place, um, you're probably in, in good shape. I mean, shameless plug. If you're unsure how to check that we offer that as we do meta tracking audits to make sure all that stuff's in place and, and dialed in, but, um, yeah, that's where you'd want to look to see if you see that multiple, like it's coming in from multiple channels and you look at that graph and you're getting server and
pixel events pretty consistently, you're probably in good shape. So meta events manager would be where you want to take a look at that. I have a pull up right now. I'm looking at it. Does automatic website matching impact, um, like conversion API or EMQ or not really? No. So, so EMQ, when we look at it and meta events manager is actually, that's all from copy.
So if you don't have a copy connection, that tab won't even show in events manager. Um, automatic advanced matching, if we distill it down is really just what you're telling the pixel, it's a pixel based tool is any page that the pixels installed on that you're saying, go ahead and scan that page in that code.
And if you see things that are email addresses, phone numbers, names, named fields, go ahead and pass that data back with the pixel for match purposes. So you're just giving it some automated leash to identify that information and send it back and that does help match rates, which we can dive into more, but that's really like how the platform optimizes.
So if I go on, um, Zach store and I make a purchase meta needs to match that action up that purchase on his site up with a user on their platform. So they need to connect that back. And then once they know I'm a user on their platform, then they're like, okay, let's go find other Brett fishes who with all of our data points has similar interests and we think would also be a prospective customer.
That's how the meta optimization engine works for ads. So it's really important that you pass as much of that, that match data back as possible to help the platform do that and more efficiently optimize your ads. So it's a pixel based thing.
The automatic advanced matching, but we highly recommend having it on and having any of the match keys on, unless you have like a strong business reason. Some of them aren't as important, like gender date of birth, probably not as important, but like name and email and external IDs and things like that are really good to pass if you can. Yeah. I mean, that's like one of the first things that we check in audits just as like to make sure it's turned on and almost always not all of this turned on.
So maybe that's like a low hanging fruit thing that people can do. The other thing I would tell anybody, I call it the silent killer and that's probably a little dramatic, but I can, I can't tell you how many audits we find issues with this. There's a section in the settings tab of events manager.
If you scroll down that says event set up tool and event set up tool is a codeless tool that meta offers. So if your pixel is installed on a site and you really need to get an event tracking on like a button click or a link click, but you don't have access to the code and you can't add events and stuff, you can use this event set up tool, which is a codeless tool for meta to do that.
The problem is, is what we see with so many advertisers is at some point in the past, too many cooks in the kitchen, summer marketing intern, whatever, has set up events via the event set up tool. And they have a normal regular e-commerce tracking set up in place. So they'll be getting like huge duplication on like add to carts or sometimes purchases and things like that. And if you look in the site code, even you won't see it, it doesn't exist.
You can only see it if you're in that event set up tool. So that's always something I recommend people check too. It's just like, make sure there's no events in there. It serves its purpose, but you can't pass back like dollar values and skews or anything else. It's just a pure like event tracking.
So, um, that's always one I recommend in the settings too, Brad, that you check on audits because it can cause all kinds of problems. So probably some intern reading the Foxville digital blog, you know, full of a lot of goodies on there, your conversation with Brett about the EMQ and about your own EMQ. What, what were you trying to solve? And then like, what is it? And what did you know about it? And then what did Brett kind of illuminate for you there? So I was curious on.
Like, I mean, I had the, I had the understanding that event match quality, the better it is, the better, you know, better results you get, right, the more data you can send back to meta, I would hope that the better performance we get out of, out of our ad account, right? So simply put, that's all I knew. Right.
What is it actually in, what was my question for Brett was like, what do the numbers mean? Right? Is a nine good, is a seven good? I have no idea. It's out of 10. So like, what are these scores actually mean? And what do I need to get worried? Right? So one of the things that we were running on one of the brands is that we were having issues, all sorts of issues with, with our, um, basically pixel tracking.
We had health and wellness hit us. A bunch of stuff started to happen. Um, that was all accidental that we got reversed. But, um, one of the things I was curious on is like this, this event, event match quality, and I was like, okay. I see the number went down and it came back up. Like, where should this be? So that was one of my first questions was like, I don't know what's a good score.
You know, sometimes you get graded. Is it, is a nine good? Is that an A or is that actually, you know, are we, you know, curving that a little bit? Like, I don't know. Right. So that was the first question was like, what do the numbers actually mean? And how good do they need to be? Do they need to be 10 out of 10 or can they be lower than that? What does that look like? So Brett, why don't you like go through that? Let's talk to you like the main events that we would see page view, add a cart, purchase the sheet checkout, like, how do you think about those?
What, what did the numbers look like? Is there a number that kind of raises a red flag for you or yeah. Tell us more about those. Yeah. So, um, so the first thing is EMQ stands for event match quality. So that's, it's important to remember because that's what we were just talking about is it's a, it's a quantitative way for meta to kind of tell you how well, how confidently can we match this event, this action on your website to a user on our platform that's, that's the gist of it. Um, so, you know, it's, it's what data are we passing back to help meta make
those matches, which drives optimization, which obviously helps your ads performance and efficiency and all that good stuff. Um, it's on a scale out of 10. I've never seen a 10 on any event with real data. So don't obsess about that to those listening. Like I've never seen an actual real 10. The other thing I would say before we kind of dive into the details is EMQ is a little funny, funky.
Um, it's weighted. It can be gamed. There's, there's a lot of elements to it. So I think what we can go through is like, what are the important things to focus on and to care about and not dwell on like it's not a 10 or it's, or it's this score, but more like what we're sending back. So, um, it, it, so the first thing is EMQ will vary.
It's always starts the lowest on like the highest funnel event. So when the user hits your website and you get like a page view event, that first landing page view, you're going to get the lowest EMQ scores of your range there, um, and we see everything there from like a four and a half to like low sixes, um, depending on what gets passed back. So for what it's worth hollow, I'll just run through my numbers.
And then we can use this as an example, seven on the dot 7.0 page view, which seems high, which is great. So I've got a 6.3 pulled up. So you're winning. And we can dive in in a minute to a little bit of what's like fueling that and why, but that's a good score. So page view 7.0 view content.
So to your point, like as you kind of get through the funnel, right? To add to cart. So page view seven view content, seven one, add to cart 7.2 purchase nine points or initiate checkout 8.5 purchase 9.0. Um, so just to give an example, live example brand that we're spending, you know, seven figures a month on, there's a lot of events going through this. So not just a few. Um, but yeah, keep going, Brett. All right. Yeah.
And the reason, but the re and the reason for that is when, when someone hits your website, they haven't given you any information. They haven't submitted anything. You don't have any of that user match data to send back. There are some technical parameters that you can capture like when they first hit. So, um, uh, Facebook's Facebook has two parameters and FVP and an FBC, which we'll dive in a little more. Cause I think the FBC is a really important one and valuable to dive in a little more on in a second, but there's those, there's like your IP address.
There's a bunch of things that you can pass when a user hits before they've submitted anything to you. That's what's going to fuel that, that page view score. Um, also a lot of stores will like, if you have a user logged in, possibly that has an account that's logged in, you might have access to their email and their phone number or something right away when they hit the site.
So those things all fuel that page view, EMC score, E and Q score right when someone hits, but they haven't made a transaction, you know, they haven't submitted like their name or their email or anything like that. So we get those technical kind of parameters. Shopify also has one. There's a parameter called an external ID, which is just like a customer ID. Um, Shopify assigns a unique one when the user hits your site and it carries through their, their whole session.
So you can pass all those things when you use your hits before they've made a transaction and sent you. So that's, that's what fuels those likes. Like we see a lot of like fives, low sixes, seven is great. You must have some like logged in users. There's another thing to mention here is there's, there's some data providers that also use like IP geolocation.
And this gets a little in the weeds, but like they look at your IP address and they guess your city, state, and country and all that based on your IP. And they pass that information back. So sometimes we go in an audit and it's like a page view and we're like, we're getting state and country and zip code on a hundred percent of events.
How's that possible? Like we don't know that yet. That's usually like IP geolocation. And that's what I mean by like, EMC can be gained a little bit. So when you see those numbers and for anyone that wants to look, if you want to look at what your EM actually makes up your EMQ score, the main overall scores will be in your events manager overview tab.
But if you expand on any of the events and go into the event details, there's a little sub tab called event quality. And you can see like the rate of all these things being captured. Um, that's just that they're being captured. It's not saying they're unique or authentic value. So I could put Brett Fish as the first and last name on every event on a site. And it would say I'm capturing first and last name a hundred percent of the time.
Although that's not valuable to the platform, right? So it can be gained a little bit, but, but that's what's fueling those like upper funnel EMQ scores. Um, the more data to exact original point, of course, the more data you send the better, as long as it's actual real authentic relevant data and not just like, I'm just obsessed with trying to drive up my EMQ score. So I'm going to send a bunch of kind of BS back to meta.
That's not, that's not helping your efforts at all. Anyway. And then you get all the way down the funnel to a purchase. I make a purchase on the site. I'm probably submitting my name, my email address, my actual address, shipping address and all that. So now we can pass all that additional information back in addition to what we had passed before.
And that's what you gets you up into the high eights, low nines. Um, typically for a purchase, which let's be honest, purchase events, probably the most important. It's what the majority of advertisers are optimizing toward. So the match rates really critical. We see usually like high eights to low nines, like nine three is usually like the highest I've seen it.
Um, high eights you're probably fine. Like eight six, eight seven, anything below an eight. Jack speaking to red flags is something we would be like, we want to at least look at this and see why. Um, metas official documentation is that a six is the minimum they recommend to actually use to optimize your ads on, but a purchase event with all the data that should be available there.
You should be in the high eights, low nines in that, in that area. So you're going on, you go through this, you're looking at this, you're grandeur listening to this and you're looking at the numbers. What are the things that you go in and would do to, to like some top things you go in and try to solve this to make these a little bit better for some folks. Right.
Or is it all over the map? It's, it's a little relative to each event. Like what you could practically hope to capture and send with each event. I think you just have to be realistic about that. Um, what we would do if like we were going to look through it is what are we receiving at each stage and does that make sense? Um, like, and, and it's, there's a little bit of common sense, not even technical.
It's like, if someone's added something to their cart, do you have their email address, probably not unless they're logged in. So am I worried, you know, if, if there's no email on an ad to cart, no, but we should be getting their IP address. We should be getting other things. Um, two, two important things I think is a good time to mention.
The two most important highest priority match keys and metas whole system. And they have documentation on this are the Facebook click ID, the FBC parameter and the email, the user email. Those are the two most important things. If you don't, if you can't send anything else, if you send those two parameters and Facebook click IDs, definitely like the top one, and I'll talk about that in just a second, but those are the two most important. So especially on a purchase event, I would really focus on those.
I'd look at like, are we getting an email? We should probably have it on a hundred percent of events. And then are we getting the Facebook click ID? Now here's the caveat with the Facebook click ID. And this came up when I was talking to Zach, it's not going to be at a hundred in your events manager.
Events manager is every event your site is receiving, not just what's being driven by ads. So even if a user originates from an organic Google search, you're going to get a purchase event fire when they purchase on your site, the click ID, that FBC parameter is only going to be present on sessions and purchases that came from a meta ad on one of their entities.
So a lot of times we'll go into a purchase and like FBC parameter will be like 27% of purchase events have the Facebook click ID. That's not a red flag. That's not a problem. What we always recommend is this. Like, if you have the data, like compare that to generally what you think that the volume of the proportion of your purchases are that come from meta ads.
So like, if, if you think that 75% of all your purchases come from a meta ad, but you're only at 20% click ID, maybe that's something to look at because there should be more of those events then that have that click ID, but it's not a red flag to see that number lower than usual in the events manager, a lower than a hundred because it's only going to be from the events that came from meta ads and that's we're looking at all events. Would, so you said compared to kind of what you think meta is driving.
So it could be, it could be higher than that in the event that maybe you have a ton of organic traffic, but meta ads are doing a solid job at capturing that and retargeting it with like a DPA or maybe some other stuff, but it might be lower than that if you have like a high AOV and the purchase, the person took 30, well, maybe you can, you could tell me it's like if they click on the ad and they went and researched and came back to it two weeks later after they saw something else, are they going to pass through is a Facebook click ID going to exist on that purchase event?
It's possible. It all comes down to the cookie expiration because what happens is, and we won't get too technical in the weeds, but what happens is when a user clicks your ad on Facebook or Instagram, meta pens, a URL parameter to that link coming through to your website, kind of like a UTM that a lot of people are familiar with, and it's the FBC LID is you can see it in the URL.
That is a really important value because what that says to meta is it's just a string value, but it says this user clicked this ad at this time and then went to your site, so it's super deterministic. So if you pass that back with your events, medic can look at that FBC and be like, well, Brett fish, click this ad at this time at this, and then made a purchase on Zach site 10 minutes after we're definitely attributing that purchase.
So it's really deterministic, but what happens is when you hit the site, if you have the meta pixel installed, it'll grab that URL parameter and store it as a cookie in the browser. And so it really just comes down to the cookie expiration, which a lot of times is like, I mean, your attribution and ads manager, a lot of people use like a seven day click attribution, but the cookie expiration is usually around like 28 days. So that value to your question, Brad should still be in there if they come back, but you know, in ads manager, if you're only using seven day attribution,
if it's passed out, it won't attribute, but that click ID will still should still be valid as long as the cookie is valid. Got it. Cool. And if you had to Google what deterministic meant when Brett said that word, like you're not alone, so it's all good. Um, I get it. It's the most important though.
And I, the reason and Zach and I talked a little bit about this on our call, it is the most important parameter because it's meta provided, right? It's not like I'm submitting an email and we hope the email that I submit on my order on the store matches my email that I'm registered with on, on Facebook and Instagram, so we can match that up. That value is coming directly from meta. They know who the user is, when they click the ad, what ad they clicked.
So it's, it's very, very deterministic. If you're going to pass one parameter, that's the one and there's, there's a lot of misunderstanding around it. Um, but it is important. So, but I always want to point out just because it's not on a hundred percent of your events, that's not a red flag or anything to freak out about.
That's normal, but that is a really important one to make sure you're capturing and passing, especially if you, um, if you happen to be an advertiser that's doing like Cappy only and not using the pixel. So you don't have that automatic mechanism of setting the cookie and all that. You want to make sure you're manually, you can still manually capture that value and pass it with Cappy. You want to make sure you're doing that.
Cause that's a really important value to your optimization and add success on the platform. Is there anything else that you guys talked about in your call that you think you don't want to dig into or do you want to switch over to the health and wellness, uh, total non-disaster? I think. Yeah. I mean, that was mostly, that was mostly it, right? I think it was just like getting the understanding of what, really how Cappy works.
It's kind of like two channels of information, trying to match some of that to like, you know, be duplicate information, ideally Cappy or server side data is even stronger to your point to give meta, even more. Data back than what you might miss, uh, that, that the pixel might miss. Um, so that clarity was helpful. The EMC, EMCQ, uh, event managed quality component was important too, because again, like without the context of like, is a nine good for purchase? You kind of freak out and you're like, maybe this is why my ads aren't, aren't working.
So I think that was just like peace of mind more than anything. FPC component of it too. Like we were looking across the multiple brands that we own and operate. And one of them was at like 48% um, FPC and purchase. And one was at like 63 and it was like, okay, that makes more sense on the 63 one, because it's very impulse purchase related.
Like I think more of our traffic is actually from meta driving that purchase than even other channels, organic email, other, other, you know, click based traffic. So to get the FPC tag to carry through makes a lot of sense. So again, I think for me, it was more just making sure everything was in good standing and that if there was something missing that we could at least identify it, so, um, which I think is really like the best rationale of like these audits with bread is like to get that peace of mind, right? Cause otherwise you might just be thinking your ads are, you know, working
great where they might not be, or you're freaking out like I was thinking that that's the issue. So, um, meta health and wellness. Let's get into that. Just before we dive into that though, is, you know, met is one of the main growth engines and channels for most advertisers and brands. These, these same things kind of about EMQ, like, um, Google has what they call enhanced conversions.
Um, tick tock has match rates, all the same kind of stuff. So this is more and more like across all major digital ad platforms is with like what medicals EMQ, but is passing back as much user data as possible so that these platforms can match website actions with their platform users. It's big across all channels.
Obviously with meta, it's very like granular and broken down and like, you can really dive in and see it, but it is relevant and important on, on a lot of these ad channels, not just meta, but metas like, you know, where most people focus and where it's like really easy to dive into your scores and all that.
So for what it's worth health and wellness, what's, uh, what's the business? What's going on? Everybody, everybody freaking out for no reason, or, uh, is it actually a problem? No, it's a problem.
Um, there's a lot of people I know that would, would definitely tell you it's a problem, uh, I can give, I mean, feel free to ask any questions like you guys have on it, but I can give kind of the quick, we were kind of like accidentally ex sort of experts in this area. Um, but I can give you the quick kind of backstory and framing. So that does sound good. Yeah, absolutely. So the real start of this is met has been hit with a lot of class action lawsuits about, about user data and privacy.
And that becomes even more. Important and critical around industries like health and finance, where that data is very, you know, obviously a lot of people have heard of HIPAA, um, health privacy data policies and things like that. Um, it becomes even more sensitive and important around things like health and finance.
So back in it's, we started getting notices, I think in December of 24. So this last year, and then things really started rolling out in January of this year, but so meta had rolled out this essentially this. Process and set up in their system where they scan the website, the URL, where you're sending all of these great events that we've been talking about, like purchase events, things like that.
When those come in, they scan that website with an automated system and kind of like how the ad review process works and how they scan ad landing pages. Similar, but different and distinct. They scan and look for, for any content or data that could be. Sensitive. So health and wellness related financial related things like that.
Um, the original rollout that happened in January, they started it, I think their highest priority level, which was anything that goes after a specific condition. So if you're selling a treatment for. Anxiety or cancer or something like, like very specific, they went after those advertisers first and what happens is they scan the URL on the website. And then they apply a category that's like a health and wellness category.
Um, if you're curious, if you are, are in that category, it might be in the events manager that we talked about in the settings, there's a section that says manage data source categories. And you can go in there and see if your domain has been categorized, but. Um, they rolled out a category that was health and wellness condition.
And there to your point, Brad, this is a big deal because the restriction that came that accompanied that in events manager was you lost your lower funnel events, most likely, um, the reason for that is those are all intent based events. Right. So anyone can accidentally click an ad hit a website, but let's say I have athletes foot, right? Gross. I go, I see an Adam like best athletes, but treatment suite, I go to the site.
I add to cart and I purchase that sending a pretty strong signal back to meta that that I have athletes foot because I purchased a treatment and that's what that site's about. So that's what they were trying to eliminate is so they eliminated everything below, lower in the funnel interview content. So I had the car initiate checkout, all the pay ad payment method purchase, all that stuff and went away.
So that's a huge deal because 90% of advertisers are, are optimizing for purchase, you lose that event. Your performance is going to pretty much start degrading immediately, which is what we saw. So, um, you know, there were some options that came up. There were options on X that were talked a lot about like, um, using custom events. Ironically, meta originally said, don't do that. They produce documentation that said, don't do that. But now there's like.
Dozens of firsthand accounts of meta reps actually telling advertisers use custom events and just keep advertising. But that was the solution then was, you know, um, use an event that in all practical purposes is a purchase event, but call it something else, call it success or call it, you know, whatever. And we'll pass that back. Only those standard events like purchase are blocked.
So you can keep using these custom events. And that was, that was a solution for a while that worked. I'd say it was probably about Mike, just my take was about 60% successful. We heard some that it didn't work for, but some it did work for.
Um, and then in June, they introduced a new wrinkle and they introduced a new categorization that was health and wellness other, which is a lot more gray area and like generalistic, but it's essentially any business that's like, or advertiser that's tangential to health and wellness. So not even a condition, but, um, you know, it can be all sorts of things that are related to health and wellness. And you get hit with this health and wellness other categorization.
And the restriction is, is that you get put into what's called core setup. You don't get lose your bottom of funnel events necessarily, but they put you into core setup. Core setup. I've seen some stuff on Twitter from some people like, oh, core setup will destroy your ad account. I don't totally agree with that on a technical level. Cause core setup was around before health and wellness.
They've used it in like other industries like finance. It's just a way to close some data loopholes so that as advertisers, we can't pass sensitive data back in like URL parameters like UTMs or in event data. So I can't add a custom parameter to my purchase event. That's like credit score, right? They block that cause they're like, you're in finance. You can't send that stuff. So it's a, it's a way to close data loopholes.
What's been interesting though is, and this is from talking to dozens and dozens of advertisers. It seems though, however, when that categorization is applied to the dataset, that there is some negative impact on the ad account. Um, one client referred to it as a mysterious force, which is, I think is about as specific as we can be.
Meta has not produced any like documentation or comms on this, but you know, once we saw it across. Dozens and dozens of accounts, you're like, okay, this isn't just a one-off. This is a, this is a trend. This is a thing where once that health and wellness categorization hits the dataset and you're put into core setup, we've seen like spend dropping conversion rate, dropping CPA, like going up, like negative impact on the ad account.
And so all we're left to sort of deduce from that is once you get categorized, there is a negative impact on your ad account and a lot of advertisers have been struggling with that. So to your point, Brad, like, I know that was like a long winded backstory, but it's a very real thing. And there's a lot of really high level operators that I respect. And I know they know their stuff where they're like, nothing else changed.
This happened and it's been like a decline since then and immediately. So it's definitely a real thing. It's impacted a lot of advertisers. Unfortunately, there hasn't been a lot of clarity and comms around like who qualifies or doesn't the only real option you have is there's a manual review button that you can be like request review.
And then you sit there like this and hope that that works. And then if it comes back and is like your SLL that's it. And so, um, there are some options. But it's, that's really all you can do. No, no great options. Um, I mean, since I've been running and he's dick pills.com, it's been tough for us. Honestly, though, there's no, a lot of your own information back to meta.
Yeah, there isn't, there doesn't seem to be a great way to navigate this, which is one of the most depressing things that I tell, like we had a rash of people that joined off of a blog post. Um, that we still regularly have people that are affected by this. And I'm like, there's really not anything magical I can tell you to do. Those that approach you, Brad, is it similar? Like you said, there's not, are there anything that they can do? I mean, you can obviously eliminate claims from your landing pages, you know, that kind of thing, but it doesn't, not anything technically. There are options. Um, there are solutions that I won't necessarily go into detail on the podcast.
Um, one of those 387 listeners might be an engineer at meta and I don't want them coming for me, but, um, there are, there are a couple of things we can do that we've helped some advertisers put in place to deal with that. Um, if anyone that's listening is interested, just reach out, um, my emails, just Brett, be our ETT at tag hero.io or send a message to our website.
But just, just reach out and like, we can talk through like what we would want to look at is exactly what are you seeing? Like what restrictions do you have? What are you facing? And then we could talk about what are some potential solutions. Um, there are some things that can be done. Um, candidly, there's nothing that the only things you can do that comply with terms, metas terms are you use your like good old view content event for optimization, which is an ideal, or you try to scrub your site and submit for a review, but like we've had clients be like, even if we could do that, like, especially in medical, they're like, it's going to take us
almost like a year between generating all the new copy, running it through legal approvals, like getting it up on the site and redesign, like it's just going to take forever. So there's not a lot of good options. And candidly, we've seen a really low success rate on those reviews. Um, the ones we've heard about that are successful usually have aggressive, um, reps at meta inside to help push that stuff through, but we see a really low rate on those. We have seen some successful, but overall low rate. So outside of those things, there are a couple of technical things that we
might be able to do to help, but it really depends on your situation and, and what you're seeing and dealing with. And then maybe there's some ways we can help, but, um, it's, it's tough for everyone because, you know, a lot of these advertisers, we, you know, you, we have a client and a friend who sells bath bombs and they were categorized and he's just like WTF, right? Like where we sell bath bombs, like, is this really health and wellness? And it's tough. And they were having, you know, some pretty significant negative impact on their ad account. So it has really impacted a lot of advertisers.
And it's one of those things that's just curve ball, right? And like very frustrating for a lot of them. Yeah. I mean, I, I was quoted in an investor business daily article that I got interviewed for about this, that I think it's fundamentally unfair that some of these people, some people can like hound a rep and can like get basically, you know, oh yeah, no worries. And then there's all these other people that exist in the world that are like running completely legitimate companies that don't have the same level.
This just feels it's just unfair. And there's always the golden rule with that too, right? If you spend enough gold, you can break the rules. So like there's elements of that too. Yeah. Yeah. Um, guys, anything else? I have, um, maybe this is like two in the weeds and maybe this is a topic for another discussion, but have you seen, um, anything work well related to not optimizing for purchase? Like you've set up, so you mentioned like the new, new customer event, but that's still kind of related to purchase. But I know there's a bunch of like quiz funnel completions you could try. I know Cody from Jones road has talked about, um, other mid funnel events that
he's testing, he's seen, he's seen some success with, he specifically mentioned they had a great podcast with, um, Eric. I don't want to say his last name, Seafruit maybe. Um, and they kind of talked about this idea of signal engineering. Something that Cody said that was really interesting. And I think Eric said on the podcast, these are try is like adding in some friction if you're going to try a mid funnel event, like, Hey, send people to the landing page, but don't fire the view content on the landing page, but optimize review content, which is the next page after that.
So somebody has to go through the landing page and then get to the product page, have you seen any like crazy setups that have actually been meaningful? Everybody's just guessing and seeing if it works. No, I mean, and at the end of the day, it's hard because I'd say like, try test it and every account is so different that like, it's rare that blanket advice just works for everybody.
Um, you know, and the other thing I'd just be upfront about is we, we, we don't always get the backend, like media buying insight and results. Like after we've worked on stuff or put things in, they don't come back to us necessarily and be like, we do it definitely here on some things like this works versus not, but like, um, I think some of that stuff is definitely worth testing, you know, I think, I think especially for large brands, like a Jones road, high spend, right? I think you get to a certain point where you can almost tap out like the purchase
event optimization. Cause if you think about it, when that is really doing is saying, well, we were talking about earlier with the MQ, Brad, you go on Jones road, you get, you know, some, some exfoliant, whatever it is. And they're like, cool, Brad bought this. Let's go find more brands. Well, you get to the point where that, where that process you start like with huge daily spends and stuff, you start to tap that out.
So I've definitely heard of large brands with large spends, like over time, especially where it's like, we've been running ads for years. We're spending a hundred thousand dollars a day or whatever. And it's like, you get to the point where maybe introducing some upper funnel events just will sort of tweak and shock the system a little bit and go after a different pool and sort of cohort of users on the platform and bring them in.
So I think like optimizing for that stuff is definitely worth a test, like optimizing for upper funnel events. And then, yeah, I mean, I, I think at the end of the day, what is a purchase event to meta, right? It's an intent. It's this user converted on the site. So we want to go find more of them. So I think in that same light, if you want to optimize for a quiz completion, that's pretty high intent.
Someone clicked your ad and went to a site and went through a whole quiz process and submitted some info. That's pretty high intent. Let's go after other people who are going to complete the quiz. Cause at the end of the day, worst case scenario, maybe they're not converting immediately, but you are getting like, they're, they're getting pixeled when they hit your site for retargeting or, you know, you're getting their email, maybe with the quiz for email marketing and retention. There's, there's some other things, positives that can come out of it too,
other than just a pure like purchase transaction conversion. So I think it's definitely worth testing that stuff. One, just like anecdotal thing we saw a long time ago when I was running more on my agency, we had a client that was very lead based. They, they produce tons of lead magnets, like recipes and things like that.
And initially we were optimizing for the lead event, right? Which was great. We got tons of leads and like we put these lead magnets out and got a lot of leads. But when we did a cohort analysis afterward, they weren't, we were getting a lot of leads and they're, you know, they might've been converting other ways, but we weren't getting actually a ton of like purchase conversions at the end of it.
Like commensurate with the lead volume. So we ran a test and instead of optimizing for leads, we actually optimized for purchases so that those ads with those lead magnets were actually shown to people who were more like our purchasers and not people who were just more likely to complete a lead and it takes a little bit of work outside of just like period platform metrics, but when we looked at the cohort analysis on the backend, that group, even though it was a lead magnet, they're just down there. The purchase value of that cohort was much higher than the lead event optimization.
So I definitely think there's an element of like signal engineering and kind of like what you want. I would still tell most advertisers optimize for what you want, which is a purchase, but I think if you get to certain size and scale or you want to try other intent based things like, like quizzes or, you know, the thing with like the view content after a landing page to add a little more friction, what that does is that makes that view content event have a little more intent to it because you had to take an extra step. Um, I know the guys that are obvious at one point, we're working on this too.
Cause like they were like, how much signal can we send the minute they hit the website? So we're capturing as much as possible. Like there's a couple of different ways to look at that, but, um, especially the stuff on the higher intense stuff like quizzes, I think it's fire an event. When you complete a quiz, optimize toward that and see what that does.
Maybe take some winning ads that you know are successful and try that and see how the results compare. Um, I think a lot of that is definitely worth testing and accounts for sure. Yeah. And we saw the same thing with lead optimization for what's worth with people who do like early access to new product launches and things like that.
Um, and then that's probably ultimately also why meta changed the way that you can optimize for a conversion based lead in more recently where you can connect your CRM and actually tell them, okay, people are actually progressing down the funnel and that you can optimize for those things, those things better because yeah, optimizing for lead might just get you a bunch of shit leads.
If you don't add some friction there. So yeah, that's a great point. Yep. Cool. Brett, thank you for joining us and, uh, we appreciate it. And, uh, if anybody has any questions, feel free to hit up tag hero or email me, Andrew at Foxwell, digital.com. Um, I've had five people do that. So wow, big numbers and, uh, everybody's one of them. Yeah.
Everybody have a great rest of the day. Thank you. This episode is brought to you by Homestead's email and SMS service. Honestly, we've, we've done an episode with, with Jacob from Homestead. He's the man and he's got, he's got to be probably one of the smartest email and retention operators like on the agency side hands down.
I look forward to seeing the tweets that he puts out basically every single month with like a recap of the designs and as amazing as the designs themselves are, it's the actual like tactical behind the scenes work. Where he just like knows what he's doing. And the team at Homestead, like really know what they're doing on email.
SMS probably, if not the number one, very, very close to competing for a number one retention agency in the DTC space. So just killing it. Yeah, absolutely agree. You know, if you are looking at your emails and you're like, these are terrible and you know, you can optimize the flows and you know, you can do SMS better. You got to look at the Homestead team. So check them out, reach out and they can definitely help make your entire email and retention and SMS program so much better.
And I'm not just saying that we've all seen the work. I actually had a member who sat down with, with Jacob at the founder meetup that we had, and he came over to me and said, I just talked to him for 10 minutes and literally he, he's revolutionized my entire email department. So agencies are modeling themselves off of what Jacob is giving them for advice.
So anyway, onwards to hopefully working with them. The only way that we grow this podcast is by you sharing it with your friends. Honestly, like reviews kind of don't really mean anything too much anymore. They're really meaningful, but they don't do a lot for the growth of the podcast.
And so sharing YouTube links, sharing Spotify links, sharing Apple, whatever we call it under the podcast app now, anything you can share, the better we're going to be guys, anything else you want to say on this? Yeah, please go check us out on YouTube. Rack up those views for us. We'd love to see it. And then subscribe. Make sure to subscribe on YouTube as well. And I relentlessly refresh the YouTube comments because it dictates my mental health for the day. So please say something nice about all of us.
Thank you everyone. Thanks for listening. Honestly.
