Heroes, Sidekicks, Villains & Zombies: Collin Slattery on Google Shopping
Collin Slattery (Taikun Digital) joins Brad and Andrew for a no-fluff breakdown of why Google Ads breaks the brain of most DTC operators who cut their teeth on Meta. The core argument lands early and lands hard: your relationship with Google is structurally adversarial in a way your relationship with Meta isn't, because Google can't manufacture more search inventory the way Meta manufactures more feed inventory. The only lever they have to grow revenue is extracting more from advertisers, which means every default setting, every recommendation, and every "AI enhancement" is built to serve them first.
From that foundation, Collin walks through the operational stack that actually works: ruthlessly segmenting brand from non-brand across both search and shopping, building a real product feed instead of plugging Shopify into Merchant Center and calling it done, structuring shopping into heroes, sidekicks, villains, and zombies, and treating YouTube Shorts as the only YouTube placement worth touching unless you're running holdout tests. Along the way he drops a study Tyken ran across roughly $20M in spend showing that hidden search terms cost advertisers about $0.85 of efficiency per dollar spent, the case for YouTube being the single most incremental channel in digital (87% net-new traffic for one of his nine-figure clients, plus 34 million Americans reachable nowhere else), and a hard truth that Google scaling is brick-by-brick, never one banger creative away.
If you've ever wondered why your Google account "kind of works but never really scales," this is the episode that explains exactly why and exactly what to do about it.
Key Takeaways
How your "5x Google ROAS" is actually just brand search dragging a broken non-brand campaign across the finish line.
How much of your Google spend is being skimmed by search terms Google won't let you see.
Why your relationship with Google is fundamentally adversarial.
How to reach 34 million potential customers that literally cannot be reached on Meta or TikTok.
If you're not running holdout tests, 90% of your YouTube dollars are likely going to the wrong placement.
The four shopping campaign buckets every brand over $1M should be using, and the one that is silently eating 80% of your budget.
Could this one setting be making your entire Google account non-incremental without you knowing?
Why Google scaling looks nothing like Meta scaling and what it actually means for your monthly growth curve.
The $50/month tool that fixes the feed quality issue almost every Shopify brand is silently losing money on.
When paying for brand search actually makes sense
This episode of the Scalability School podcast is sponsored by NorthBeam and they just launched Northbeam Incrementality. Northbeam Incrementality gives you easy, automated, self-service incrementality tests, while protecting you from the major mistakes so many people make while running incrementality tests. Your MTA handles the daily tactics, your MMM guides the long-term planning, and Incrementality provides the causal truth. It’s a closed loop that allows you to scale what works and cut what doesn't. Right now when you head over to www.northbeam.io/incrementality, they’re offering Scalability School listeners 50% off unlimited tests for a year when you join. Just tell them we sent you!
To connect with Collin, send him a DM at https://x.com/CJSlattery or at: https://taikundigital.com/
To connect with Andrew Foxwell send an email Andrew@foxwelldigital.com
To connect with Brad Ploch send him a DM at https://x.com/brad_ploch
To connect with Zach Stuck send him a DM at https://x.com/zachmstuck
Learn More about the Foxwell Founders Community at https://foxwellfounders.com
Learn More about the The Hive Haus Creators Community at http://HiveHausUGC.com
Full Transcript
Clean Transcript
I thought it would be interesting too. I know Zach, I was looking back on this tweet from like almost a year. It's almost a year ago now. It was actually June of last year. It feels like it was just yesterday that Zach was shitting on all Google ads agencies. But I thought this was a good starting point.
And so Zach's tweet, I'll probably summarize this. But Zach kind of started by saying, if you're a DTC brand, being your Google ads agency more than 10K a month, you are very likely getting robbed. Zach is known to take strong positions. Usually with good reason. Usually with good reason. But sometimes he has strong takes.
And so we don't have to spend a bunch of time dispelling this. But just curious if you have a kind of quick take on your feeling of, I guess, maybe the average Google ad agency and what you guys do differently that helps set you apart. Well, obviously, I think that's bullshit. And I said as much around that time and the engagement was fun. And I know Zach loves to throw bombs out there and then turn around and walk away. But yeah, no, I mean, that's clearly not true.
Right. And I actually would say that Google in particular is one of the places where investment in top quality talent is even more paramount because of the difficulty and the number of dark patterns that exist in the platform. This gets to a foundational concept around the differences between Google and Meta. And one of those is that the relationship that you have with Google is inherently adversarial, where it is not necessarily adversarial with Meta. And the reason it's adversarial with Google is because Google is a monopoly. And they don't really have a direct impact on search query volume.
Right. Like the amount of people who are going to search for tennis shoes in any given month is relatively set. Right. Like they can't really impact that. And they already have the full market. They can't get any more market share.
And so the only thing that they can do and speaking primarily around search and shopping and things like that is increase the amount of money that they get paid for the queries that already exist. That's the only lever that they have to pull to increase how much money they make. They're a publicly traded company. They're in the business of making money. Right. And when they are increasing the costs, that is born by the advertiser.
Meta, on the other hand, has the ability to impact their inventory in a way that Google does not. Right. They can increase the frequency of ads in reels or your feed. They can make the product more addictive. So people scroll deeper so they can be exposed to more ads. Right.
And so fundamentally, Google is not playing nicely with advertisers because it is in their best interest to. Take money from advertisers. And so if you have someone who doesn't know what they're doing and doesn't know Google's dark patterns, it is very easy for Google to be an absolute waste of money because that's good for them. Right. Because they need to make money. So if you don't have someone who's excellent, and admittedly, there are not that many excellent people out there, it's going to be a non-incremental waste fest.
And I think that's what most people run into because they're working with people who don't know how to do it or they're working with people who are meta first and the rules between meta and Google are completely different. And then they're like, oh, Google's trash. Everyone who does Google is trash. And screw this. I'm known for taking notes. And the first thing I wrote down is Zach equals skill issue.
So that's what I heard from section number one here. It is, in fact, a skill issue. A hundred percent. Good. Good. Well, I'm sure Zach will hear this and we'll remind him.
But that segues nicely into like, OK, so then what can brands that are listening to this do and take away to actually start to make Google a channel that is effective for them? Because I think a lot of people will. I mean, our average, the average client that we work with has Google and there's brand set up in some capacity. Whether or not it's incremental, there's a bunch of questions around that. And yeah, right. And you mentioned you mentioned there's there's probably different things to consider, certainly across industries and how you should consider across industries, but even account spend and size.
And so I think if you could just start to kind of take people through like what what is the actual account setup and structure look like? How does it change as you scale? I think that could be a really interesting starting point to just kick off, you know, where people can start to get some tactical takeaways. Yeah, and I think this is this gets back to the size and complexity of Google, which is material. It's a materially more complex platform than meta is. There's a lot more stuff going on and there's a lot of stuff going on that.
Is beneficial to Google, but detrimental to the advertiser. And I think if people are. Looking to succeed on Google, I think that is a mindset shift that needs to happen where when you are in there, the. Everything you look at, you need to think this is set up in order to benefit Google and not benefit me. So you should be very skeptical of anything that Google recommends. Right.
And we're kind of that way with meta when they're like auto apply all of these AI enhancements and shit like that. But those are much more visible in meta than they are in Google. And so you absolutely have to be approaching it with that mindset and trying to make it harder for Google to succeed. And so brand is like this stereotypical example. And we see it with brands. Of all sizes, even from, you know, seven figure brands to 10 figure brands, you go into their Google ads account and it's very rare that brand and non brand is truly fully segmented apart.
And it has to be because if you say I want a five X right. And you put brand in there, right, that's going to generate a 20 X. And then Google is like, cool, we got a 20. We banked the 20 X here. That means we can waste money on stuff that doesn't perform. It'll average out to a five.
And the advertiser is happy because it says a five. But they haven't dug that layer below and seen all those conversions are happening to brand. 80% of the budget is 10% of the conversion value and 10% of the budget is 90% of the conversion value. And blended, it looks great. They run a holdout test on it. Like, hey, this is not incremental at all. Like we chopped our Google and nothing happened.
It's like, well, yeah, of course. It's because your brand and non brand are blended together. And so if there is one thing that people take away, get them split apart completely, right? And split in search. So negative it out in all your non brands. If you want to run it, put in its own manual CPC search campaign.
But you have to do it on shopping too, right? Like shopping inventory serves against brand queries. And so if you're not segmenting out branded queries and shopping, if you don't have a branded shopping campaign, all that's going to roll into your Pmax is going to roll into your regular shopping. And it's going to be the same problem that you would have with search as Justin Shopping. That is, I think, the biggest problem and also the easiest problem to fix. Is that a one-time thing?
Or do you have to check it regularly? Because Google is trying to jam brand back into things or brand adjacent things. So if you do a good job with your negative keyword list, you're going to get 99% of it out. It's not going to cover all misspellings. And so it does make sense so much in a while go through and negative out the misspellings. When you launch new campaigns, you do need to apply the negative keyword list to the campaign.
But it's mostly a one-time, 30 minutes max sort of activity. So there's a lot of brands. Brad was saying there's a lot of small brands or smaller brands that listen to this podcast. And I think, you know, have always wanted to test Google. And I think people get excited about brand. They get excited about launching Google and they get excited about launching YouTube ads as well.
So when you're looking at those brands that you've worked with in the last, let's just say 12 months, because I know it changes over time. What are they doing? What's the mix of brand, non-brand? Are they doing shopping? Are they doing YouTube? What does that mix look like?
And what are the things that you've seen be truly incremental to their results and their growth? I'm going to say something, and maybe this is my first hot take, even though it's not a hot take because it's driven by data. Google and YouTube done well are the most incremental advertising that any brand will ever have, period, full stop, done. We've done this with nine-figure brands where 87% of the traffic coming from YouTube is net new to the brands, whereas Meta is sitting around like 60, 62%. Google and YouTube are so incremental and expose so many people to traffic that is nowhere else. There are 34 million Americans that are not on Meta properties, are not on TikTok, are not in any other digital properties, but YouTube.
Literally, the only place you can reach these 34 million people are on YouTube. Totally net new, do not exist anywhere else. They're only on YouTube. And you can reach them there, right? And so, and that's just the people that aren't anywhere else, right? It's not even counting the people that you don't reach because they're not in a conversion-focused audience in Meta or anything like that.
So, massively, massively, massively, massively incremental. It's just set up poorly most of the time. Now, to kind of like loop back and it's like, okay, well, how do you make Google incremental? Well, one is, it's get brand and non-brand separated, right? And then what you're doing is different based on the scale of the brand and where your budget allocation is. The first thing I'd say is like, don't run brand at all.
Like negative it out and then don't have it running, period. And I think Zach will probably agree with me on that, right? It's generally a waste of money. And it's one of the very first holdout tests that we run with any client that comes on, whether it's Meta in Google or just Google only, is like, let's test brand, like see if it's incremental because it's usually not. And it's something you can just turn off, which is great, right? If you can allocate that budget to things, put it as a YouTube shorts instead.
Like what a better way to do it. So if you can get away with not running brand, don't run brand. And then the structure is like, it's going to be pretty consistent, but the budget allocations will be different. So non-brand shopping or like a non-brand shopping Pmax, right? Non-brand search. And then if you are running brand, brand search, brand shopping, and then some degree of demand gen shorts as kind of like the third component of that.
So, you know, we're looking at three to five campaigns as kind of like the default structure. When do you, so then, okay, so that's for everybody. So what is scaling it out look like? Is it horizontally adding more campaign by product, product category? Is it just scaling it up by cranking budgets and sending it to the moon? Obviously, YouTube is probably handled fairly differently because there's a creative component to it, whether that's shorts or just, you know, general YouTube placements.
So you can pick either side, maybe Google is the place to start and then work over to YouTube. But like, what is actually, what does it look like to start scaling up the Google side? Yeah. So I think the easiest scale to unlock is shopping, right? It's the shopping inventory. And it's easy, but it's not simple because what dictates how shopping works is really the structured data that you're giving Google, right?
So it's the quality of your product feed. It's the descriptions. It's the images. It's the secondary images. The color data, the size data. It's how optimized your titles are, right?
It's making sure that there aren't price mismatches or any other issues that are impacting the quality of the data that is being fed to Google because that's how Google decides, oh, this is, this is, you know, a blue sneaker, right? It's, well, you have to give them all that data. And so if someone searches for blue sneaker, right, you want to show up for that. And so the only way you're going to show up is if you have good, clean data. And reviews are another huge component of that. People are going to go, they're going to search for blue sneakers.
What's going to get surfaced and what's going to get clicked on. Stuff that has good data so that Google knows that's what you're selling. And then things that have a lot of reviews because that's a trust signal and people are going to click on it. And so feeding Google good data is going to impact how much shopping volume you're going to be able to acquire. And then there are other parts that obviously brands can't control. So like if you, the larger your SKU library, the greater the surface area you can cover.
And so generally the higher the scale that you can conceivably get. Similarly to that, the better your business economics are, right? If you think about search impression share, like a pie or a hundred percent of search impression share, like no business is going to be able to economically monetize a hundred percent of search impression share. But if your economics are not so good, right? If your revenue per session is generally pretty low or your margins are pretty low, like the percentage of the search impressions where you can profitably target it, it's going to shrink. And so your business metrics also like dictate kind of the scale that you can achieve on shopping as well.
But there's a ton of scale in shopping. So, I mean, it's interesting, you know, I've never heard somebody give the case for shopping that heavily, so that's interesting. I think you've had this thing about called hidden search terms and you have a framework of Google inventory, which is premium, mid-tier, and garbage. And that's what, this is, you know, taken from your word. So talk about hidden search terms and the research around this and what things folks can do to bring some of these search terms out that maybe they weren't, you know, aware, ever familiar with in terms of targeting. And this is, again, this gets to the whole, there's additional layers underneath Google that when you just look at it, it's painting a very small picture of what's actually going on. so keywords and search terms are not even remotely the same thing, right?
It's like, oh, well, you know, I'm targeting blue shoes. Like, that's the keyword that I'm targeting. That's what I'm bidding. It's just like, no, that can match to hundreds of different search queries that are somewhat related to completely unrelated depending on your match type, right? And so those are the search terms that surface. And so those are the things, search terms are what people are actually typing into Google to find and then your ad is surfing against, right?
And Google, up until, I think it was probably 2020, like 2021, I can't remember the exact year. At this point, it was a while ago. They used to show 95% of all of the terms that people typed in to, that surfaced your ad. So you'd have 95% visibility, 90 to 95% visibility in everything that people searched for that your ad surfed against. And for privacy reasons, they decided to hide a huge percentage of the search terms that people searched for that your ad surfed against. And they lumped them together in what is known as other search terms, which you can't see what they searched for.
You can only see a single line of the aggregated performance of what is hidden. And that's a problem, right? Like you're paying for it. You should know what people are searching. And the performance of the stuff that is hidden is generally worse than what is visible. And that's because in my estimation, right, it behooves Google to hide the stuff that people would try to exclude if it's not related.
Right? If my keyword is, you know, blue shoes and, you know, the matching is blue tablecloth, like, that's not related. Right? But you can get matches that are that far away. It's actually pretty shocking how not close search terms can be to the keyword that triggered them. And now that they can hide those, it's easier for you to keep bidding on them as opposed to historically, you know, turn them into negatives and remove them.
Yeah. You put together like a pretty gnarly report, like actually digging into this. Do you have any of the numbers like handy out of the top of your head that you can just like rip off? Or is that a little bit too specific? I mean, do you want, I have the report somewhere probably. Pull it up.
Yeah. If it's not, if it's not a huge pain to pull up, I'd be super curious to dig into that because I remember reading it I'm pretty sure that, is that what Ginny responded to? And I do want to note this is only for search campaigns and shopping campaigns where you have full search career report data available. But we looked at around $20 million in ad spent in search and shopping spent and found that for every dollar of ad spent that you are running through Google, they're skimming almost a full dollar. off of your efficiency with hidden search terms. So, if you were to spend a million dollars on ads and everything was visible, you would generate about $5.6 million in revenue. In reality, you're only generating about $4.75 million in revenue.
And that $850,000 shortfall is due to the inefficiency of hidden search terms. Yeah. So you're looking at like a 5.6x for visible versus a 4.75 for full blend including hidden. Like worst case scenario, historically, before this change, you could just exclude them and, okay, I've excluded them, I'm not wasting money on them. And now, it's just being lumped in there. And so, what you're saying is best case scenario, you can actually exclude them and then the additional spend, you'd be able to continue.
You're spending that money anyway, so it's going to be spent at that more efficient return number. Am I understanding that correctly? Well, I mean, you can't get rid of the stuff that's hidden, right? For sure. Stuff is always going to be hidden and it is a game of whack-a-mole to consistently exclude things to try and pull up efficiency numbers. You can aggregate data through what's known as Ngram analysis to pull stuff out of hidden, like, and exclude stuff that would otherwise be lumped into hidden search terms.
Generally, hidden search terms are stuff that Google categorizes as, like, low volume, right? So it's not getting a lot of volume against those queries and so those are generally going to be longer tail search queries and by longer tail, I mean, more words in the query. And so you can aggregate what is visible and find terms that are low performing and when you exclude those, it will also generally pull out stuff that was hidden that also included that term that is not effective. Not usually going to get like a lower percentage of hidden but you can reduce the inefficiency overall by going through and pulling stuff out. Cool. Personally, I'm very interested in like the actual tactical levers of, okay, we can dig into shopping.
Andrew, if you're cool with going down that route, I'm just going to start ripping some questions. Absolutely. Feel free to take it wherever direction you want to. So you mentioned you've kind of got like four to five campaigns that are standard. I think we'll ignore the brand side of this completely for now because I think your basic TLDR is like turn it off or highly limit the spend that's going into it at all if you're going to spend into it and just be thoughtful around that. If you're in a vertical where you need to spend on brand, you'll know because it's the exception.
Fair. Cool. Maybe worth coming back to that because that's interesting. But then shopping is where you're seeing the majority of the volume. You said every brand could spend six figures. It's kind of like a take.
So as people are thinking about setting up shopping, you said the biggest lever is kind of just like the inputs to Google which is functionally the feed and maybe some offshoots of that. So what are brands doing wrong when they set up their shopping? You mentioned the title and the actual data is not set up properly. So let's say that they are starting to chip away at updating that and maybe their 80% of what you call like this is a good input. The feed is solid. What's like what's still missing?
And then if they wanted to increase spend volume through shopping, like what are they actually doing? It is actually I should say is one of the biggest challenges with Google is because everything is different depending on small changes in a brand, right? Like if you have 10,000 SKUs, like your approach is going to be completely different than if you have 50 or 100 SKUs or 10 SKUs, right? And so the feed is table stakes, right? And having that fully optimized and I will say like most people just connect their Shopify feed to Google Merchant Center and they're like, I'm done, right? Like, no.
And if that is you, like, go get data feed watch. It's not that expensive. It's going to allow you to really optimize your Google fields, which is super important. And we're going to assume you have reviews piping in because that's also very important. Shopping from a structural level and I think Google in general, and this is something that we've seen time and time again on search, on shopping, absolutely on YouTube. Google is not nearly as efficient a budget allocator as Meta.
And so, so many people come from the Meta space where it's like, trust Meta, they know what they're doing, go broad, like drop it all together, right? You cannot do that with Google. You're going to have a bad time if you're doing that with Google. And that extends to shopping as well. And then you need segmentation and you need thoughtful segmentation in order to unlock volume. What this segmentation looks like is really dependent on number of SKUs and business priorities.
If you're a retailer and you carry a hundred brands, your shopping structure is going to look materially different than if you're a supplement company and you've got 20 SKUs. Just try the example, though, Colin, of let's say there's this lower amount of SKUs, 20 SKUs. How are they structured? Are they structured based on which one's driving the best revenue for you with the best gross margin? How do you start to structure that? I think the initial structuring that you're going to want to do, and again, this is a phase one, is you have one campaign and you have ad groups in the shopping campaign that are segmented either by a category or if you have really very few products, you can even segment them individually.
So you're a supplement company, you have supplements for digestion and supplements for sleep, right? You can break them up into those individual categories and then make sure that you have negative keywords at the ad group level to make sure that you're routing queries to the correct ad groups. So that's kind of like a we're just getting started structure. And then once you have that running, you're going to start to see, okay, this product is potentially working really well and performing really well. Maybe that's something where you actually can, and we've done this, like we'll have a shopping campaign with just a single product in it so that it's not competing for budget with the other products in that campaign. It's not super common where we'll fully pull out an individual product, but it'll happen.
Over time, breaking the campaigns apart based on by performance and like bucket them together really into like four main structures which are like high spend, high performing, high performing, low spend, just low performing, and then like no spend, right? And so, you know, you don't want like your top 10% of SKUs to be competing for budget with the 80% of SKUs that get 2% of the total clicks, right? It's historically called, you know, heroes, sidekicks, villains, and zombies, right? It's kind of the four monikers for the different structures. And you break them all apart. And SKUs can shift in between the different campaigns, but the reasoning behind that is Google's budget allocation is not incredibly efficient.
And if you're lumping them together, you're just leaving money on the table. But the first segmentation starts in an individual campaign at kind of the accurate level by grouping products separately. And grouping, does grouping the products on its own give to your point? Because it helps Google sort out delivery a bit better because you're inserting kind of an opinion into how that should be structured. Does that alone add more volume? Or is it, hey, we need to split these things up and then we go edit product titles to have more, you know, like, are we trying to increase the search volume by getting more keywords?
Like, I'm like, very tactical as somebody who's not running Google accounts. Like, so you're segmenting those things, but like, what else is the team doing on a weekly basis to try and drive more volume? Yeah, so I mean, you're also removing inefficient search terms, right? that's just an ongoing negative keyword application is an ongoing piece of work in Google, on shopping, on search, everywhere. Product optimization, title testing, that is also a key component of it because, again, like, you operate under the understanding that there is a finite amount of search volume for whatever it is that you're doing, right? And when you find things that work, you want to be getting a high, as high a click share as you possibly can. So optimizing to try and maximize your click share, generally, like, it's through titles, it's through more reviews, price, obviously, also a component to it, using promotional things, promos to get the sort of slash through and, like, sale tags and things like that.
And then adjusting targeting, right? And that's one of the added benefits there, too, is, like, if you have things broken up, you can have a specific target for, you know, mascara, whereas blush has a completely different target and you can control those efficiency targets individually. Yeah. What role do images play in shopping? Like, are you guys testing images consistently or is it not a big lever? Generally, like, pretty strict limitations around images in Google. rules around, like, not having texts on the images, like, not putting promotional information on images in the feed.
So most things do default to white box. You can include lifestyle images as well. There are fields for lifestyle images that always worth doing. Basically, and this is another general rule, if there's an optional field that you can use, it's not optional. Just use it, right? Like, the more optional things you include, the better your feed score is going to be in Google's estimation and the more opportunity you have to surface.
So there's not a ton that you can do from an image standpoint in terms of testing. You can do a little bit, but it's tough. We've had clients where their product is in a box and the box has text on it and they'll reject the image because of the amount of text that's on the box. I hate to host this phrase, but it's kind of phrased this way intentionally, but like, is there anything hacky in shopping that you can do that consistently works? Like, you know, there's this theme on Meta for a while where everybody did the we're sorry kind of ad in that positioning. Like, there's a format that pops up that you can just run that playbook across brands and it works very consistently.
Is there any comparable on the Google side, whether it's like hacky feeling? Maybe it's specific to shopping, maybe not. Not really. And I would say that in general, like, it's not how we like to approach advertising. Whatever arbitrage exists, I mean, you're opening yourself up to potential issues down the road. You know, like, there was a time where you could like, get promo text onto images by, you know, making it less obvious and tricking the sort of machine to think it wasn't there, even though it clearly was.
And you're just going to get your feet suspended and you're going to have to redo all the work. So, you know, I generally operate on the do things the right way principle and you're going to be in good shape. Brad's always trying to hack an idea, you know. Look, I'm just, you know, you never know what exists to me. I'm the anti-hack guy. I'm the let's do it by the book and, you know, always be on the right side.
I also lean that direction, but, you know, sometimes people like, you know, a little sauce every once in a while. Do not invite the eye of Sauron, you know. That's fair. So let's just talk about YouTube briefly in going through that, right? I think this is a place where people are doing a ton with taking stuff from meta, putting it on shorts. What do you see that works, that drives incremental results for folks and how do you think they should approach YouTube?
How much time do we have? YouTube is the most complex place to advertise and the hardest place to advertise in digital. And YouTube is like six or seven different things, right? So like when we talk about YouTube, I mean, we're talking about so many different things. And so, you know, I'll start with kind of the the low, lowest of low-hanging fruit, which is that YouTube shorts thing. It's the demand gen conversion focused YouTube shorts component where you can take your high-performing 9x16 video assets from meta, put them in a demand gen campaign, and run it purely on YouTube shorts placement.
You know, get kind of click based, pretty good click based attribution out of it. you do have to do audience targeting. That's one of the things to remember. You know, I always say like YouTube today is kind of like meta in 2015 where, you know, you do the 1% lookalike audience, right? Like we were on this call all old enough to remember the old lookalike audience stacks and things like that. So you do have to do audience targeting there. But that's kind of like the lowest form of YouTube advertising.
Do you have any issues with the measurement on shorts? Like, is it cool? Like, do you trust what YouTube is reporting on shorts enough? Or, like, I'm thinking of the brand that's not paying house. I'm sure house is fantastic. I know they're fantastic.
But like, just a quick sidebar on measurement of shorts. It sounds like it's low-hanging fruit, but what about the measurement side? So, shorts measurement is okay, but not amazing. And Google is, it's always going to be under counting the efficacy of shorts. And I will say this very declaratively. If you don't have the ability to do holdout testing, do not spend a dollar on YouTube outside of shorts.
Shorts is the only place you can spend money. Do not do in feed, do not do in stream, do not do CTV, do not do any of it. Like, if you are not doing holdout testing, don't spend money on sort of traditional YouTube. Just only spend money on shorts because that's the place where you're going to get some click-based attribution and the rest of the YouTube inventory. And this is why YouTube inventory is so mispriced, right? It's like, it is hard to measure.
You do really need to be doing those holdout tests. So, stay in shorts and don't venture out of the shallow waters into the deep end of YouTube. I mean, I think it's a matter, I think that's good about the measurement and making sure that you have a plan for measurement and makes a ton of sense. To me, it's one of these things that you hear people talking about. They don't really know how to set it up, right? I think a lot of people spend a lot of time optimizing their meta ads accounts.
And I think YouTube is a place that takes a lot of energy and attention and people just want to repurpose stuff, which you can do to a degree, but I guess I'm just curious of any other tips and tricks and tactics that you believe are really valuable. The other big thing is audience, right? Audience still matters on YouTube. And Google's targeting on YouTube is the best and most granular targeting that exists today in digital. Like, you really can target your ICP. And Google's bidding algorithm is not at the level where you can just go broad and it will find your ICP, right?
So, you know, if you sell beauty products, like, an in-market audience for beauty products is a good audience to test on YouTube, right? Like, go narrow. Go top 10% household income women between the ages of 35 and 54 who have X, Y, and Z interests, right? Like, the kind of stuff that we used to do on Meta, like, still totally applicable on Google. You should definitely be doing that sort of targeting with YouTube, even on shorts, right? And, like, shorts is huge.
Like, it's bigger than reels. It's bigger than TikTok. Two billion people use shorts a month. Like, shorts is colossal. So the inventory is there for essentially any advertiser to spend thousands of dollars a day on just shorts as a placement. But, again, you gotta take what's old is new again.
You gotta take the old rules of what you did on Meta, you know, back in the day and get granular with your audiences. And take your winners, right? Like, you know, I think short form video that performs well is short form video that performs well. You know, obviously, like, TikTok's kind of like the gold standard. Like, if it performs on TikTok, it'll perform everywhere else. Like, it performs on Meta, it might not necessarily perform on TikTok.
But take your winners, drop them into a demand gen shorts campaign, have some granular audience targeting with different ad groups and doing granular audience targeting. And, like, that's a super, super easy place to start. It will have, like, pretty good click attribution to it. And, you know, you're not going to spend $20,000 a day on it, but $2,000, $3,000, $4,000, $5,000 a day, I mean, it's not nothing. I'm very curious about the role of landing pages across Google. I know from a shopping side, it would have to be handled differently because it needs to go to functionally a PDP.
So, I guess you could split it into two parts. Like, what are the role of landing pages from a search perspective? I'm assuming you have a lot more flexibility in what you can and can't do from a search side. But, to your point earlier, maybe there's a little bit less volume overall in search. But then I'm just curious, like, have you seen, like, CRO-specific tests to PDPs that have been impactful for Google traffic specifically? Like, obviously, price and offer is, you know, maybe not where I'm going with this.
But I'm just curious if you've come across any, like, CRO wins that have unlocked crazy scale, whether it is a shopping or search side. Absolutely. Yeah. Yeah. And landing page is huge. You know, landing pages are being important on Google, like, predates them being important on Meta.
I mean, it's historically been very important because landing page quality score is one of the components of quality score. And quality score impacts your CPCs. And so a better, higher converting landing page will increase your quality score, which means it helps your ad rank, which means your CPCs are going to be lower. So it's always been a key component to it. And you can run shopping to landing pages. And especially if you are in, you know, maybe something that does not have a lot of SKUs, if you're in kind of like a, you're a supplement with like a massive hero product, you probably don't want to be sending that to a PDP.
You do want to be sending that to a landing page that is specifically structured for that traffic. Because again, we get back to the how can we maximize the inventory that we can target profitably, right? Like that's going to impact your revenue per session, right? And that means you can adjust your targets and you're going to be able to get a greater click share, right? Like I would love to have a 90% click share for search terms that are good for me, right? Because that's what's going to increase my volume.
So landing pages are massively critical to search, but also to shopping. And I, you know, I'm not saying create a custom landing page for every single product that you have, but if it gets a lot of volume and you look and you see, okay, my click share is relatively low. Like, okay, well, how am I increasing that? How am I increasing the click share for this specific product? Like, let's make a landing page for it because now our revenue per session goes up, which means we can, you know, adjust our bids, which means the percentage of the inventory that we can get and get profitably is going to go up. So, yeah, landing page is massively critical.
It's the key to unlocking additional non-brand search as well because people are looking for specific things and your products might have 20 different angles for what it does, right? Like a magnesium supplement, like it's great for sleep, but it's also like good for focus and all these other things. And if you're sending all of those queries to your PDP, that's terrible, right? It's like best supplement for sleep. All right, well, there you go. Now we're hitting our sleep queries, right?
Supplement for focus. Now we're hitting something completely different. And so, you know, and this is, I think, the biggest challenge with Google is that outside of YouTube, there's not one big unlock. You're not going to find a killer creative and scale your budgets 10x. Like that's not how it works. It is brick by brick incremental layering of a few percentage points of efficiency here.
Unlocking a new class of keywords that's going to add another, you know, $3,000, $5,000, $10,000 of ad spend a month. That's running efficiently, right? You're layering on these bricks and building it one piece at a time. There's no, hey, we found an absolute banger, like, let's rip it to the moon. Like, that's just not how it works. And, you know, I think that lends itself to a specific type of person who is very in the weeds, who is going to do all of this work.
And also why I think most people who deal with Google or most brands are like, oh, Google doesn't work or, you know, it's because they're not willing to pay people who can actually do this to build it and build it over time because it does layer over time. It is not, we took it from $10,000 a month to $400,000 a month in, you know, two months with one banger ad. It's layered over time. I think we have a new podcast called Skill Issue that can come out of this to bring this full circle where we just like, we'll just pull up ad accounts and Colin, you can just publicly shit on everybody's Google account. It's so hard. There are not that many Google people in the world who are good at it.
It is, and I've seen this with clients who have been wanting to fire us for a very long time, not because we're not doing a great job, but just because we're expensive and they're like, you're too expensive. We don't want to pay you. We're going to in-house this. I'm like, okay, good luck. There are just not that many people who are legitimately good at Google because the number of years it takes to be really good at Google is a lot and the attention to detail is massive and the number of individual settings that if done incorrectly will completely invalidate everything else that you did is incredibly high. You can do all of the things that I've said so far through here completely correctly, but if you have your conversion set up incorrectly, which is very common, let's say you have two different primary conversions, it's going to be completely non-incremental.
It's going to generate zero revenue and it's going to look great, but it's so easy to have page view plus purchase as primary conversions and they're the account defaults and so they're there in that campaign and you're telling Google, I value page views and I value purchases and so obviously one is much easier to get than the other and so Google is going to just be like generating traffic that's firing page views and it's going to be racking up like efficiency in there, but you're not going to be seeing it on your bottom line and it's because that one single setting was done incorrectly and now it's all worthless, right? And there's dozens of those throughout the account and you need someone who can be completely on top of that, knows where they are and isn't going to make a mistake, right? And there's just it's hard work and there's not a lot of people who have that depth of knowledge to do it in a platform and then you add on YouTube, which is a totally different platform within the platform and it's like, okay.
