Google Ads Success Secrets: What’s Working Now in ads for eCommerce with John Horn

John Horn has worked on over 1000 Google Ad accounts and is here to give us the low down. 

Here's what we cover: 

• Why Performance Max is king - but only with guardrails. 
• How Standard shopping still has a place. 
• What Google Demand Gen Campaigns are, and why they are the new frontier.

Plus we discuss how 3rd Party Attribution can sometimes be more of a hindrance than a help.

Transcript

Jessie Healy Hi everyone, and welcome back to the e-commerce impact podcast. I am your host, Jesse Healy. And today we have John Horn  here to talk to us about Google Ads. Super excited about this episode. I have so many questions that I want to dig into, and I have been doing Google Ads for nearly 20 years, but it changes all the time.

Jessie Healy And so one thing I know about Google Ads is you've got to talk to the people that are managing budgets and doing lots of, lots of good stuff on there and able to kind of tell you what's working now. So that's what we're going to do today. So John. Welcome along. And why don't you introduce yourself to the audience and let us know a little bit about you and what brought you to where you are today.

John Horn Jesse, thank you so much for having me. I'm excited to chat Google Ads and e-comm today. [00:01:00] And yeah, like you mentioned, I'm the CEO of Stub Group. We're a digital advertising agency, really focused on paid search, paid social and we work with a wide variety of brands. We worked with over 2000 clients over time and have helped generate over half a billion dollars in revenue through Google Ads and other PPC platforms.

Jessie Healy Wow. 2000 clients. Did you say

John Horn That's right. 

Jessie Healy That is quite a lot of experience, I must say. So yeah, that's awesome. Okay, cool. So my first question to you is, what is working now in Google Ads? Let's maybe deal with the brand, like, let's talk about the brands and maybe two groups, like if they're an earlier stage brand, and then maybe a bigger brand.

Jessie Healy So starting with early stage. What sort of set up are you going for? What, what works best for e-commerce brands who are wanting to, you know, scale up their, their revenue and sell more product food through Google. 

John Horn Yeah. Let's let's dig in. So There's a lot of places I could go with this. I guess I think I'll start with talking about types of campaigns and then we can go [00:02:00] from there. So the types of campaigns that are kind of options right now for eCommerce prone, generally speaking, are search campaigns where you're showing text ads for people who are searching on Google.

John Horn com and search partners. You've got shopping campaigns which display the shopping listing ads with price and product image and things like that on Google. Then you've got Performance Max, which I'm sure we'll talk about a lot today. And that's Google's kind of one stop shop campaign that shows ads across shopping and search and display and YouTube and so forth. And then you've got higher funnel campaigns like demand gen, which shows mostly on YouTube and display. And you've got actual YouTube campaigns, actual display campaigns. So for a lot of brands right now especially if they're just getting into Google Ads, or even if they've been doing Google Ads for a while, a lot of the questions surround, okay, well, there's all these, there's all these different campaign types.

John Horn Like what should I be using right now? And as usual with Google, the answer is always, it depends. But a lot of the framework we're looking at right now comes down to how much history they do or don't have [00:03:00] with Google Ads and that will kind of dictate what we start with. So, if you've got a brand that's just getting into Google Ads, hasn't done it before, they're obviously not going to have history in their account.

John Horn They're not going to have conversion data history either. Conversion history is really important these days because with Google's bidding strategies, which, of course, is how they decide or how they take your input and then decide how much to bid for different traffic and where to show you, the best bidding strategies that work these days generally speaking in e-commerce are what they call their smart bidding strategies which are based upon conversion history but if you have no conversion history, then Google has nothing to base that upon. So for kind of brand new brands or brands that are just getting into Google Ads, generally, we'll start with shopping ads and potentially search ads, depending upon their budget and how large their audience is. The reason we do that is because there's more control and more insight into what's actually working with those campaign types. So if we run [00:04:00] shopping ads, I can actually look and see what are the search terms that are triggering our ads. And get a sense for, is Google showing us the right traffic? Do we need to sculpt traffic, add negative keywords, so that we're getting, we're not wasting money for the wrong traffic?

John Horn And kind of, try and hone things in, to where we're getting the right people, the right audience. And start building some conversion history. Now, if we're talking about a brand who already has a lot of conversion history, especially a brand who's maybe already tapping into some of the low hanging fruit through shopping campaigns and search campaigns, then a Performance Max is definitely something we're going to consider.

John Horn And for those listening or watching, you don't know what Performance Max is, it's a type of campaign that Google has been heavily, heavily pushing over the last couple of years since they brought it out. And it's very, very AI machine learning based where there are a lot of levers that Google has control over that you don't as the advertiser. I will say Google is giving us more control over time because of a backlash from advertisers who don't like the black [00:05:00] box. So there's actually some positives there, but by and large, you're kind of giving your assets. You're giving your product feed and things like that to Google. You're giving your target return on ad spend or cost per conversion. And then Google is going and making informed decisions to try and accomplish those goals. And if you have a good amount of the conversion data, ideally, if you have at least a hundred conversions per month and when I say conversion, usually we're talking about an order placed on the website coming from the campaigns.

John Horn That's what I would call a conversion from an e-comm brand. Then Google can start kind of optimizing with that and and going, going from there. And so for a lot of brands, I kind of see that, that structure right now.

Jessie Healy Cool. So when we're talking about that early stage brand that doesn't have much conversion data and you said shopping plus the other one was, was it brand or was it search?

John Horn The search. Yeah.

John Horn Shopping and Search. 

Jessie Healy Okay. And so is shopping a shopping campaign actually still an option or we're actually sort of hacking performance max and just doing a, like a feed only campaign, which is effectively a shopping campaign.

Jessie Healy [00:06:00] How's the actual set up.

John Horn Great question. It is actually still possible to run shopping by itself. They call them standard shopping campaigns.

John Horn What Google got rid of, they used to have what they also called smart shopping campaigns and that kind of got co-opted by Performance Max, but you can still run a standard shopping. And Google actually rolled out a somewhat surprising update just this past month, where they clarified that if you have a standard shopping campaign running at the same time as a Performance Max campaign running, and they're both advertising the same product, then instead of historically, Google would give Performance Max preference to show it now, whichever campaign has the higher ad rank, which in this case is bid. If you had the same products in each campaign, that's what will dictate which campaign serves the ad. So Google's got a little bit of an about face where it looked like standard shopping was going away. And now it's actually. Got a little more, a little more love from Google at the moment.

Jessie Healy Interesting. And like Performance Max campaigns include shopping in them. So what's the reason why we can't trust the algorithm, trust [00:07:00] Google to just run the shopping. If shopping works best for us, why won't it just run shopping campaigns? Why do we have to like sort of meddle with the machine in this way by like taking away that option and running standard shopping,

John Horn Great question. Often Performance Max actually does a really good job. We have a lot of clients where we only run Performance Max, or I should say we don't run shopping, we just run Performance Max and other campaign types. There's a couple of places though, where I think standard shopping shines. One is early on for campaigns because Performance Max

John Horn Is like, you can't do manual bidding with Performance Max. You can only do target return on ad spend or target cost per acquisition bidding. And if you don't have conversions, then Google is just kind of making up guesses as to what's relevant. Whereas with standard shopping, I could say, okay, I, you know, here's my manual bids.

John Horn Let's start with this. Let's build the conversion data. So it shines there. I've also seen instances where for whatever weird, algorithmic reason with Google Performance Max actually performs better when standard shopping is also running at the same time. It's like a, a guardrail of some kind and I, I [00:08:00] don't know exactly why but there's enough people who've seen this thing happen as well that I know it's I'm not crazy so for some accounts, we'll, we'll do that especially if we're having issues with Performance Max performing the way that we want to

John Horn but ultimately I think yeah, go ahead

Jessie Healy No, that's interesting. So you're saying that sometimes Performance Max on its own, doesn't have the data it needs or the inputs it needs. Whereas running a shopping campaign alongside it can somehow sort of. Give it the, the data it needs or something that makes it work or for whatever reason it works and, and you run with it.

John Horn Exactly. Yes, I've seen that. I've seen that happen. And then the other challenge with Performance Max campaigns is a lot of people correctly believe that Performance Max skews heavily towards remarketing. So people who've already been to your website, Performance Max turns that into sales attributed to performance max, but it's actually remarketing.

John Horn And so it makes Performance Max look really good because of that display component where [00:09:00] people are maybe clicking through a display ad and you an image ad on a website and converting. And it looks like you had this amazing ROAS as from PMax Performance Max, but it's, it's skewed by that remarketing traffic and then also by brand traffic.

John Horn So they're also known to capture a lot of traffic from people searching for the brand name and then clicking, and so I'm not at all anti PMax. There's a lot of places where I see it working well, but there are a lot of those factors and kind of nuances you have to take into account when you're looking at the overall account structure.

Jessie Healy Totally. So let's dig into then a more mature brand. Let's assume the brand is doing, I don't know, maybe mid seven figures. They've got some good brand awareness. Maybe they've got some product in stores that people come and look for. How are you setting up your campaigns and how are you sort of navigating that challenge, that PMax?

Jessie Healy Confronts us with where you're not really sure whether it's preaching to the converted, it's all mopping up the bottom of funnel, or is it actually driving incremental revenue? How do you go about kind of experimenting [00:10:00] with that?

John Horn Yeah, a couple of approaches. So, on the Performance Max side in particular, there are some guardrails that we can add. So, for example, we can opt out of branded traffic, and we can set search themes, which are kind of like keywords. They're not keywords, but they're basically us telling Google and the AI, Hey, someone's searching for this type of thing is relevant to us.

John Horn And so trying to feed almost like you might prompt chat GPT to help it figure things out. You're kind of prompting Google to say this is relevant to us.

John Horn So there's, there's some different things you can structure around that. But I'd say also, as more mature brands looking to scale, then we're really starting to look at other campaign types in addition to Performance Max, because Performance Max is kind of in that low to middle spot in the funnel. It's still, you know, it's called Performance Max. It's still more performance oriented. And as we try and scale and increase brand awareness and top of funnel, then we're going to go to places like demand generation or demand gen campaigns, which are really YouTube and Google display [00:11:00] oriented, and they're more about, generating demand, about getting in front of the right people, creating awareness, getting people searching, which will then often convert in those lower funnel campaigns like Performance Max or search. So we'll do demand gen or just straight up kind of high level YouTube campaigns, which can be very, very. powerful. If we're identifying the right audience and using some really good video content the, the, yeah,

Jessie Healy So tell me more about these demand gen campaigns. That's a campaign type that you're selecting within the ads manager, like within Google Ads, it's called ads manager, Google. Yeah.

John Horn yeah, Google Ads manager that

John Horn that works. 

Jessie Healy Yeah. Yeah. Well, so within Google Ads manager, that's a campaign type that you're selecting and how does it work?

Jessie Healy Like what are you feeding it with? How does the optimization work and, and what, and what, what is, what sort of brands does it work well for?

John Horn Yeah. So it's very, very creative driven. The way Google kind of positions it as they'll often say, like, Hey, you've got stuff working well on [00:12:00] Facebook or Instagram. All right, bring it over to your demand generation campaign. So kind of think of it in that way to some extent. In terms of where the ads serve, it's going to be on places like, like YouTube and Google's display network, which is, you know, tons and tons of websites that have image ads on them.

John Horn I believe they also serve on Google's discover kind of app and platform as well.

John Horn And so create is very important again, video and and imagery. From a targeting perspective, one of the things I like about Demand Gen campaigns is that it's the only campaign type Google currently offers where you can use lookalike targeting. 

John Horn So a lot of people are aware of lookalike, you know, obviously like, you know, on Meta where you upload your list of customers and you tell Meta, go find me other people like this. You used to be able to do that for a lot of campaign types on Google, and then they really restricted that. And now Demand Gen is the only place where you can do that.

John Horn So that's, that's one thing I love is. Upload some good data and tell, tell Google to go find other similar people

Jessie Healy Awesome.

John Horn Another tactic that I'm seeing [00:13:00] work pretty well at times right now too is you can create, you know, audience is based upon a lot of different factors in Demand Gen campaigns. One of those factors is you could load in, for example, the names of competitors and their websites. And essentially you're telling Google, look, if someone is searching for this, or you know that they're spending time on this website, they're very relevant to me. And so you create an audience of those people, essentially it's a competitive audience, but not quite like on, on Google search where you're targeting people right when they're searching for your competitors name. Instead of saying, we know these people are interested in competitors.

John Horn We're going to show YouTube ads to them. We're going to show image ads to them, and that can be very powerful as well. 

Jessie Healy Yeah. Amazing. So Google knows that people have been on Lululemon's website and you can tell Google to show, so you're a leggings brand, you can tell people to go on, sorry, you can tell Google to advertise to everyone who's kind of been on similar sites to Lululemon and get ads in front of them. So presuming you know what sort of sites your customer is likely [00:14:00] to visit, you can really like, hone in on who's seeing your ads.

Jessie Healy That's awesome. So Demand Gen is like top of funnel going towards prospecting. It's, you're not using that for like retargeting, whereas we're thinking about Performance Max is like mid to bottom of funnel. And so when we're, so one thing to clarify on that, so when they brought it out, Google kind of positioned Performance Max more as like a full funnel, right?

Jessie Healy So if they kind of stepped away from that and sort of admitting that it's bottom funnel and it's Demand Gen sort of like the other piece of that puzzle?

John Horn That’s exactly right. Yeah. They've, they've stepped that back to more of, or I should say less of it will do everything and more of it's going to try and drive performance for you and mid to bottom and then Demand Gen's kind of that higher level solution.

Jessie Healy Interesting. So though at least you’re being honest with us now, hey?

John Horn Well, never go too far, but

John Horn hopefully more on 

Jessie Healy Yeah, yeah, yeah. And so if you're [00:15:00] thinking about Demand Gen, Like you run Meta Ads as well, I presume for your clients. So you're kind of running both, both platforms. How do you think about allocating budget to something like demand gen versus something like metal or even TikTok, which tends to be quite good at like demand generation introduction of a brand or a.

Jessie Healy A concept of a product to someone. Are you just allocating some budget to demand gen and comparing the performance to Meta, or are you seeing it more as like a scale strategy where it's, you only do it when Meta is sort of maxed out and you need new avenues?

John Horn I'd say a little bit more of that latter example,

John Horn simply because we've seen, we've seen inconsistency with performance from Demand Gen.

John Horn I've seen it work well at times, and I've seen it not work well at times. And so, if we're kind of thinking about our funnel, and generally the approach we'll take clients in, we would might start with, you know, we talked about search and standard shopping, and then Performance Max, and then probably from there go to Meta, and really try and tap out the low [00:16:00] hanging fruit there before we transition to Demand Gen.

John Horn Most of the time, you know, unless it's a client who really wants to test everything and is very gung-ho and just wants to, you know, have a lot of things going, in which case we'll, we'll certainly work with that.

Jessie Healy Yeah, I guess it's always that thing of like, if it's not working on Meta, then it's probably not going to work on a, on a newer, less proven platform like Demand Gen, but if you've kind of cracked Meta and you've maxed it out, then it makes sense to then try and add additional channels for further reach and scale, right?

John Horn Yeah, that's right.

Jessie Healy Yeah, that's how I would think about it too. Okay. So you mentioned with Performance Max that you can exclude brand and you could potentially run like a non brand PMax and then I guess a branded PMax at the same time. Is that typically what you're doing for your clients? Cause I know there's a lot of debate in the industry and I know there's no one right answer.

Jessie Healy Do you have one kind of approach or do you like do some, you know, test both for brands? Like how do you think about the brand [00:17:00] versus non brand?

John Horn it's definitely nuanced across our portfolio. We don't have a, we don't have a standard. It really depends. So it depends on how much brand traffic there is and how well known the brand is or isn't. You know, if it's a newer brands just getting started, we don't really need to exclude brand traffic because they're just getting started.

John Horn If it's a brand that has a tremendous amount of brand traffic, then we're much more likely to, to separate things.

John Horn But we've also had situations where we've, where we've done that and then the client has actually asked us to reverse course on that because they're using, let's say a third party attribution tool, like a Northbeam or something like that.

John Horn And they were frankly so used to seeing the ROAS from their campaigns that included brand that they're not really comfortable with seeing much lower ROAS. Now that is truly cold traffic. And so like, Oh no, we need to, we need to keep, you know, keep doing what we're doing, even in that way. Maybe it's not the best approach from a media efficiency ratio perspective. 

John Horn So 

Jessie Healy Yeah. So it's basically vanity that makes [00:18:00] them want to put it back 

John Horn Sometimes that's, that's definitely a thing.

Jessie Healy Yeah. So, I mean, it's an interesting thing, right? So if you're thinking about the combined, like, is the reason they want to do that, that the combined ROAS is actually lower when you've split them out? Or is it that they just want to see it all in one kind of, like, one row.

Jessie Healy Because if the combined ROAS is better 

John Horn right. 

Jessie Healy yeah,

John Horn In that, in that one example, it's, it's really come to mind. I think the approach we took there is, is, Hey, you're doing so well organically right now already for your brand that, and you're spending so much money on brand and traffic that we think we can improve your, your media efficiency ratio by really de-investing in branded.

John Horn And so we took that approach. But they, they got a little bit scared because they're used to a certain amount of traffic coming from Google Ads and, you know, certain return on ad spend, which was much higher when those branded traffic was included in those campaigns. And so it was more of, okay, Hey, you know, [00:19:00] I understand that maybe branded isn't doing as well as everything else.

John Horn It's not as important, but we still need to see these metrics. And so kind of went back to, went back to the status quo in that particular example

Jessie Healy yeah, I have a many a story like that. So I worked for a very big, well known brand in the US which will remain nameless. And my, the CMO,  we wanted to test like do some holdout tests of like, we were spending a million, I think it was a million a year,

Jessie Healy I think it was a million a year on brand and we wanted to just test in some geos, like What would happen if we reduce that budget or turned it down? But the CMO was like, no, because then, you know, the CEO will see how much of our traffic actually comes from branded and we, he might cut our budgets. And it's like, I mean, we're talking about like a multi, that was a hundreds of millions of dollar brand.

Jessie Healy So sometimes I think if the brand is very corporate, you end up people like serving themselves rather than serving the brand like ultimate growth growth goals. Very [00:20:00] frustrating. So you mentioned marketing efficiency ratio. So is that like, that's probably something interesting to dig into. Do you want to explain to the audience what that exactly means and how you think about the performance of your campaigns?

Jessie Healy on that basis and how, you know, how do you tell if something's working by using marketing efficiency ratio or media efficiency ratio?

John Horn Absolutely. So the, this concept is pretty straightforward as a formula. Essentially, it's looking at, okay, what's our top line revenue, the money that we're making, and then how much are we investing into marketing? 

John Horn And we don't want to just look at, at one channel. We want to look at across the board channels and, you know, fees you're paying to agencies and things like that.

John Horn It's really, what are we investing into marketing? And then you divide revenue by that marketing cost. And that's your media efficiency ratio.

John Horn Now, what do we do with that? It's not like there's a hey, here's a good ratio and here's a bad ratio because it completely depends upon your goals. And it's also generally something that we want to look at how it's changing over time.

John Horn So

John Horn for most of our clients, we're looking [00:21:00] at on a monthly basis and looking, you know, month to month, year over year. So we're looking at, okay, is revenue going up or not, and it's the efficiency of our spend, how is that impacting that? So if we're looking at last year and we're like, hey you know, just make an example Let's say a brand brought in a million dollars in revenue last October and they spend 100,000 in marketing that's a media efficiency ratio of 10

John Horn and then we look at it this year and we're like well We spent you know, 30 or 300,000 in marketing and we made 1.1 million. Well, our ratio just really dropped the efficiency of that spend as it translated into revenue, probably not a good thing. We're probably being less efficient. Let's look at what's different year over year. What's causing that? And so this is really important because a lot of people, you know, they look at return on ad spend and there's a place for return on ad spend. But they're usually when you're looking at return on ad spend, you're looking at like this. directly attributed return on ad spend. You're looking at, okay, I've got my Google Ads campaigns running. I've got my Google Ads [00:22:00] tracking code on my website. It says I generated X amount of revenue. I spent X, this is my ROAS. And that's important to be aware of to look at and important when you're kind of comparing different, similar campaigns within Google and saying, Oh, well, okay, this search campaigns got a better ROAS and that search campaign. I should probably invest more money here. But what it doesn't really take into account is how everything is combining as a whole to drive people to your website, because it's a very, as you know, very convoluted customer journey.

John Horn Someone might see a YouTube ad, then they might see your Facebook ad, and they might search organically, whatever the case. There's many different touch points possible. And so if I'm going to a client and I'm saying, hey, we should invest money into YouTube, for example, not going to see very good ROAS directly attributed for that money.

John Horn Most people are not going to click the YouTube ad right away and buy. But I'm going to watch that media efficiency ratio. And if I see, Hey, we started investing money last month into YouTube. Our top line revenue is growing and we're still, you know, relatively as efficient, maybe a little bit less efficient, but we're making a lot more revenue. That's, that's a good [00:23:00] thing. Let's see how we can keep scaling that. And so that's, that's something a lot of brands unfortunately don't spend much time looking at, but we have found it really helpful with the brands we work with when we can kind of convince them to look at it and measure it on a, on a regular basis with us, it can, it can really help make better informed decisions.

Jessie Healy Yeah. Interesting. So so if you are, you gave an example there of , kind of looking year on year inside a month, say if you're running Meta Ads and Google Ads for a brand and like you're doing an experiment over here on Google, or you've pushed some extra budget into Meta and there's maybe three or four things happening, some new creative that's launched, how are you then able to like link that to the media efficiency ratio and actually say like which action was driving.

Jessie Healy Has driven you forward or backwards or whether it was successful.

John Horn The honest answer is it's tough because it's so, it's so directional. Like, you know, if you have multiple variables at play, it can be difficult to say which of those variables impacted it, which is why I would say [00:24:00] from a testing perspective, it's more helpful when you're making like, bigger swings of we weren't doing YouTube.

John Horn Now we're spending 20,000 a month on YouTube. That's a big swing. That's going to have an impact on certainly on what our spend is. And so let's see what the impact is on revenue. If it's more of, Hey, we're trying this ad versus that ad, that kind of thing. Generally speaking, I'm still going to be looking at in app attribution, looking at row as things like that.

John Horn Trying to look at those apples to apples within the platforms to get a sense for how those things are converting, because it's, it's hard to tell, honestly, from a media efficiency ratio perspective.

Jessie Healy Yeah. And then you've got sort of like the in between option, which is using an attribution tool like Northbeam or Triple Whale. So how do you guys think about that in the agency? Like which attribution models do you think work best and like, how do you, how do they play a part in your decision making like for your in platform media buying?

John Horn These are controversial [00:25:00] waters to wade into, which I love. Everybody, I feel like everybody has a perspective on this and a strong perspective.

John Horn Here's, here's where I'm at right now. I've used a number of different third party attribution tools, examples, like you mentioned. Often, often it'll be, Hey, a client comes to us, they're already using this one.

John Horn So they want us to use it and layer into it and leverage it. And. I have a lot of skepticism at this point on how helpful a lot of those tools are because I see them kind of make very definitive proclamations that clients then take as truth, of, this is our return on ad spend from Facebook, this is our return on ad spend from Google, for example. And yet I'll kind of dig into things with them. Like a great example one of those tools, you know, a client was using them, and they're saying. And then they told us all our, you know, our reps at this, at this company said that when we look at Google performance, we need to look at click and view attribution.

John Horn And when we look at Meta performance, we should only look at view attribution. And I'm like, why? Like, like those are 

John Horn like, of course [00:26:00] Meta looks amazing right now when you're only looking at view attribution 'cause you've got so much remarketing going on, it looks amazing. And then if you layer on click attribution, suddenly that ROAS, you know, drops by 10x.

John Horn Like why would we use you know, different standards for Meta versus Google. So I think there's value in them. Because you know, they, they do, I think, help you identify some things that it's hard to identify when you're just looking at the platforms or just, you know, something as high level as media efficiency ratio. But the massive danger I see is that because it's putting hard numbers on these things that I don't think it can actually know what those hard numbers are truthfully, then The marketing manager, the CEO, whatever. Okay, boom. That's dead and sewn. That's true. And all of our decisions are gonna be based upon those things. And then, you know, you start getting into, into troubling waters.

Jessie Healy Yeah. So for instance, I have a client at the moment who we have two pieces of information. One is we turned off, we, we pulled back on Meta for various reasons by like 50 [00:27:00] percent for a month. And everything tanked, like the site traffic tanked, the email traffic tanked, direct traffic tanked, revenue tanked, everything.

Jessie Healy We turned it, we've started to turn it back on. But Triple Whale is saying that Meta has a efficiency, sorry, a ROAS of like less than one. So, and that's based on almost every attribution model apart from the view one. Now the Meta account is optimized for for view and click attribution. So that could be part of the problem, right?

Jessie Healy But it's just really, a really interesting conundrum because on a media efficiency ratio basis, like. Something bad happened when we turned off those ads, like everything else, you know, that's a pretty clear signal, right? But the attribution tools are saying that Meta is just not really doing much for the brand.

Jessie Healy So it's really tricky to know where to turn. We can't, I don't think, like you say, it's super nuanced and it's super, we have to keep doing like more [00:28:00] tests and look at both and like see how they relate to each other and then kind of decide from there. And I feel like I'm in the same camp as you. It's like, There's not just no one source of truth.

Jessie Healy These tools can be really useful because they crunch information and pull in from different sources and allow you to see, for instance, they're really great at like reporting media efficiency ratio in one place, which is actually quite hard to do, you know, when you've got Shopify here and you've got your ad spend here and you've got to pull it into a spreadsheet.

Jessie Healy So that alone is like a great thing. But yeah, I think I agree with you. Like, How are we to know inside someone's brain what actually made them buy at the end of the day? Was it because they had a talk to their friend at the school gates about how great this product is? Was it because the ad flashed up a minute before they bought the product?

Jessie Healy We don't know that. 

John Horn And there’s so many, so many gaps now too in tracking for any platform. You know, there's so much between ad blockers between, you know, the browsers who don't allow cookies between switching through devices, where there's just, it's just, you know, the reason [00:29:00] that Google Ads doesn't have perfect attribution, it's going to be the exact same reason why any other third platform is not as well, because there's just some things that it's, it's impossible to track.

John Horn And there's always going to be these gaps and correlations that exist.

Jessie Healy yeah, totally. So going back to Google now, like, do you think, here's a question for you, do you think Google's harder now than it was five years ago or easier?

John Horn That's a great question. I haven't, I haven't thought of it in that framing recently. There's so many, some aspects are harder and some are easier. So I'm thinking through from that, from that. A big picture perspective because Google has made a lot of things easier. There's a lot of things that used to be very manual, you know, a lot of our task as advertisers is button pushers of, Oh, push this percent up, that percent down, do that thing. And generally speaking, AI machine learning is better at making those decisions and analyzing things and then pushing those buttons itself than we are, so that's taken a lot of annoying hard work out of, out of the place. The flip side to that though, is that because there are fewer [00:30:00] levers to control, It can be hard to get consistent results when Performance Max does really well last month and then it tanks this month and the client says, why did it tank? It can be very hard to answer that question because there's so much of it as a black box, like, well, looks like Google's showing our ads, less on shopping network and more on display this month. Why? I don't know. Maybe it's because of this and that. And so it can be really hard conversations. So I'm going to say, you know, it's, it's the same level of complexity I'm going to say as five years ago.

John Horn It's just that there are a lot of different tasks that advertisers have to do now versus five years ago.

Jessie Healy Yeah. Yeah. It's more about like sort of dancing with the machine and trying to figure out what works rather than before. It was more a case of like creating a machine of your own that you had full control over. Yeah. 

John Horn Very good way to put it. 

Jessie Healy That kind of like skags approach. We had all these like super complicated ad sets ad groups.

Jessie Healy Sorry. Yeah. Okay. [00:31:00] Awesome. Love that. Awesome. So what other advice do you have for people kind of  navigating Google with their agency, even themselves, like what would be your top, like two or three tips for how to kind of maximize the performance of your ads right now? In 2024 nearly 2025?

John Horn I would say go back to the basics for a lot of brands. And what I mean by that is a lot of brands get so caught up in a lot of what we’re talking about right now. Like what’s the right campaign type to run, or the attribution model and whatnot, and while those things are very important, if you don’t have an amazing product, an amazing offer, if your website is not clearly communicating those things, if you don’t have credibility elements, your checkout process isn’t super seamlined, you are going to miss a lot and lot of revenue. And so while yes, a lot of those more kind of fun, flashy things are important, just- just get it right from a product fit standpoint and serve your customers well and get them returning and frankly that in and of itself puts you above a lot of the competition right now. 

Jessie Healy Yeah, yeah. It’s a really good point. I think even five years ago when there was less competition, you could get away with like a less than well optimized site, and now that’s kind of much more important and it means that you could fit a data to the algorithm to optimize from and then the machine kind of works better. 

Jessie Healy  Yeah. Totally agree with that. Awesome! well, thank you for coming along today. I learned a lot myself from this discussion and I'm sure our listeners did as well. Where can we keep in touch with you, follow you, find out more about what you're, what you're doing?

John Horn  Yeah. Two places, stubgroup. com see lots of info on there and our YouTube channel search stub group on YouTube, put a lot of free content out there as well.

Jessie Healy Amazing. Thanks for coming along.

John Horn  Jesse, thanks so much for having me.