How Chris Ridley of Evoluted Built the Award-Winning PPC Campaign That Boosted ROI 27x

Episode Description

In this episode of Automation Layering Masterclass, Chris Ridley shares the story of building Evoluted’s Search Cannibalization Management Tool—a solution designed to stop paid ads from stealing traffic from strong organic rankings while maximizing incremental growth. He explains how the tool integrates SEO Monitor’s API with Google Sheets and Google Ads scripts, ensuring ads only run when they’re truly needed.

Episode Takeaways

This discussion between Fred and Chris centers on Evoluted’s innovative Search Cannibalization Management Tool, a solution that prevents paid ads from competing against strong organic rankings while ensuring ads only run when they provide incremental value.

It spans technical architecture, human oversight requirements, future-proofing strategies, and the broader implications for the evolution of search marketing automation.

1. The problem: Search cannibalization

Search cannibalization happens when paid ads show up for keywords a brand already ranks highly for in organic search, meaning the brand might end up paying for clicks it could have gotten for free.

This was exactly the concern for Puck Stop, a client of Chris Ridley at Evoluted.

“One of our clients, Puck Stop, wanted to run Google Ads but they already had a really strong organic search presence and they were afraid of Google Ads stealing their traffic that they’ve worked hard on and we wanted to reassure them through automation that we won’t let this happen.” explains Chris.

The problem gets worse when rankings drop suddenly. Manual fixes come too late and performance suffers. Ridley shared that his team often paused ads manually, but delays in spotting ranking dips meant lost traffic.

Not all SEO-PPC overlap is bad though; if a brand ranks #1 with low competition, pausing ads can make sense. But when rankings fall below position 3, clicks drop and ads become essential. That’s why decisions should be based on real-time data, not fixed rules. Solving this also requires coordination across PPC, SEO, and dev teams, which is why Ridley’s team built an automation that keeps everything aligned and ensures ad spend drives true incremental growth.

2. Technical implementation and automation layering

To solve the search cannibalization problem efficiently, Chris Ridley’s team built a smart but simple system using tools they already had access to.

Instead of creating a brand-new platform, they connected three tools:

• SEO Monitor (to track keyword rankings)
• Google Sheets (to process the data)
• Google Ads (to take action)

This setup allowed them to automate decisions without needing a heavy development build.

“By hooking up their free API to a Google Sheet, we were able to take a daily picture snapshot of our organic rankings. And there we noticed a pattern where from looking at the SERPs as well, if their organic ranking dropped below two or three, they’re actually below the fold.

We came up with a solution to start using organic rankings to determine when our ads would turn on and off and whether we wanted to turn search ads and shopping ads off or just search.” says Chris.

The automation runs daily. SEO data is pulled each morning, and a script updates Google Ads shortly after.

Since Google Ads can’t take in SEO data directly, labels are used to link rankings to ad actions. Google Sheets acts as the hub, making it easy to manage, debug, and collaborate. Timestamps help monitor if things break, and bulk editing via Google Ads Editor keeps it scalable. It’s a low-tech, flexible setup that gets the job done without needing a custom build.

3. Human-in-the-loop approach

While automation plays a big role in the system, it’s not fully hands-off. Chris Ridley’s team tried using AI tools to match keywords automatically, but found that the tools often misunderstood context—like confusing “hockey skates” for roller or field hockey instead of ice hockey.

“The reason we’re doing manual here is we’ve tried to use LLMs and other automated ways of matching language and we found that it can get it wrong. An LLM or generative AI might assume it is like roller skates and it’s like field hockey in that way which can then make the mistake because it doesn’t understand the context.” Chris said

Another reason for human involvement is the gap between how SEO and PPC teams talk about keywords. The same business goal can look completely different in each system, so humans are needed to translate between them and apply judgment based on strategy, not just word matching. Even seemingly simple things, like word order, can trip up automated tools—so it’s safer to let people do the final checks.

Rather than rolling automation out all at once, the team recommends starting small, monitoring results, and scaling only when it works.

They also designed the system with flexibility in mind. Manual tweaks are encouraged to tailor it for each campaign’s context. Whether it’s choosing the right keywords to apply labels to or integrating new SEO tools, human input remains critical for quality, accuracy, and adaptability. In the end, the automation works best as a helper, while humans stay in charge of the strategy and fine-tuning.

4. First-party data integration

Most marketers think of first-party data as things like customer emails or purchase history—but Chris Ridley points out that SEO ranking data is also first-party data, and one that’s often overlooked. Companies collect it, but rarely use it to inform other channels like PPC. That’s a missed opportunity

“Your SEO data is also first-party data, right? So if you can connect that into Google Ads, that is yet one more way that you can set up a little bit of an advantage against everyone who’s not necessarily doing this and using that factor as a decision point in how they run PPC ads.” explains Fred.

Ridley’s system turns SEO rankings into real-time signals that guide PPC decisions, giving them a speed and strategy advantage over competitors still relying on manual reports.

Over time, this creates a unique data-driven setup that’s hard to replicate. The real value of first-party data isn’t just having it—it’s using it quickly and across channels, which most companies still don’t do.

5. Performance results and validation

The results from Ridley’s automation approach show just how powerful smart, well-timed strategy can be, even on a small budget. With just a few thousand pounds in ad spend, his team helped their client, Puck Stop, achieve a 113% increase in search revenue and an impressive 27x return on ad spend. This wasn’t about throwing more money at ads, it was about using data intelligently to run ads only when they were truly needed.

“We actually won three awards at the Northern Digital Awards event in 2024 around December time, including silver for Best Digital Marketing Campaign Retail, gold for Best Marketing Campaign B2C, and we also got Best Integrated Campaign because it involved dev, SEO and PPC.

We saw 113% growth in search revenue. And the budget we put behind this wasn’t big. It was only a couple of thousand pounds in media spend. We got a really healthy 27x return on ad spend, which was a lot more than the client expected. Conversion rates increase for the entire website by 11.66% as well. So it wasn’t just a Google Ads success story. This is a success story across all of our departments working with Puck Stop.” said Chris.

Even the client was surprised by how well the system performed, highlighting that many still underestimate what automation can achieve. Most importantly, this success proves that you don’t need a massive budget to see big results. With the right setup, automation can create lasting, scalable improvements that benefit the entire marketing operation.

6. Future-proofing for AI search evolution

Ridley’s team built their automation system with the future in mind.

Instead of tying everything to how search works today, they made the setup modular—so if search behavior shifts, like it’s starting to with AI-generated answers, only the input layer needs to change. The rest of the system, including how it decides when to show or pause ads, stays intact. That means if traditional keyword rankings become less relevant, the system can adapt to new types of signals, like visibility in AI overviews or topic-based rankings.

They also used Google Ads labels, which are flexible enough to survive changes in how campaigns are structured. So even if keywords eventually disappear as a targeting method, the same logic can be applied to ad groups or campaigns using labels. This gives the setup a kind of built-in resilience against platform changes.

“Because it’s used in labels, you could change it from a keyword label to an ad group label, you could change it to campaign label. So even if keyword come obsolete in Google Ads, you can then just take this to a different asset that you can apply labels to. Like you said, it’s completely modular.

I think Google Ads is already priming itself for that by doing topics and not allowing you to choose keywords. I think audience targeting will become a bigger thing as well.” Chris explained

Search is shifting toward more conversational, AI-driven experiences, making it harder to tie ad performance to specific keywords. Ridley’s open-source tool is designed to evolve with these changes, allowing others to adapt it as search behavior and data sources continue to shift. It’s a flexible framework built to stay relevant as the search landscape changes.

7. Geographic and scale considerations

Ridley’s system is built to work across different cities and countries, recognizing that a brand’s organic performance can vary by location. Instead of applying the same rules everywhere, the setup allows for multiple, separate automations—one per market—so each region gets tailored treatment. This is especially useful for businesses that perform well in one city but face tougher competition in another.

Scaling the system, however, comes with practical challenges. While the tech can handle it, managing large accounts means more manual setup, especially when labeling keywords correctly for each market. That’s why Ridley recommends starting small, testing the impact, and expanding gradually. Organized naming conventions and clear segmentation are key to keeping the system manageable across geographies.

Ultimately, the system is designed for flexible growth, but businesses still need to weigh where automation makes the most sense. Not every market may justify the effort, and larger expansions may require trade-offs between precision and efficiency.

8. Performance Max and advanced campaign management

Performance Max campaigns present a major challenge for automation because they don’t use traditional keywords, making it impossible to apply the same label-based logic that worked in Search campaigns. To adapt, Ridley’s team built an alternative system using negative keyword lists to control what not to show, since direct control isn’t possible. While this workaround helps, it’s far less precise and can have side effects, like unintentionally blocking shopping ads if exclusions are applied at the account level.

Not everyone has equal access to these features either; some advanced options require support from a Google Ads rep, putting smaller advertisers at a disadvantage.

“When it comes to Performance Max, this is really hard because there’s no keywords in there for you to apply labels to. There’s no keywords there for you to control.

So our solution there is by building an automated negative keyword list for your Performance Max campaign.” - Chris Ridley

Ridley’s team is actively developing new tools to address these limitations. As Google moves toward more black-box automation, the role of advertisers is shifting—from direct campaign control to influence and constraint. Automation systems will need to evolve too, focusing less on micromanagement and more on shaping outcomes within Google’s increasingly closed system.

Tools & Resources

• Evoluted’s Search Cannibalization Management Tool
• A Step-by-Step Tutorial to Using the Tool:

Episode Transcript

Frederick Vallaeys: Hello and welcome to another episode of Automation Layering Masterclass. My name is Frederick Vallaeys. I’m going to be your host today. I’m also CEO and co-founder of Optmyzr. Now for today’s episode, we have the honor of having Chris Ridley on the call. Chris runs Evoluted and he is a big fan of technology, automation, and automation layering because he wants to keep humans in the loop.

Today we’re going to bring Chris on the show and he’s going to run us through one of the ways that he’s made his life a little bit better and his PPC performance a lot better by using some of the principles of automation layering. So with that, let’s welcome Chris to the show.

Chris Ridley: Hi, thank you, Fred. Thanks for having me.

Frederick Vallaeys: Yeah, thanks for being on, Chris. You and I met in person in Manchester for the Performance MCR event. So it was great to meet you there. Over drinks we were talking and you said you had some cool stuff you were working on. So thanks for coming on and showing that to the whole audience here today.

Chris Ridley: Yeah, absolutely. It’s a tool that we’ve been working on in the background at Evoluted for a while. We first debuted it at Hero Conference UK in April and really excited to share more about it. Yeah, it was really good seeing you in Manchester as well. So yeah, really good keynote by the way.

Frederick Vallaeys: Oh, thank you. I think that might be online if people want to watch it. But besides building some cool automation layering, you also keep quite busy with twin daughters, I believe. Recent father. So congratulations on that.

Chris Ridley: Yeah, thank you very much. Yeah, they’re just learning to crawl at the moment, so nowhere is safe in the house.

Frederick Vallaeys: Just by way of intro, what’s another cool fact about you that people might not know yet, but might find interesting?

Chris Ridley: Yeah, so one of the not so cool facts about me is I broke my elbow just two weekends ago, so that’s still recovering. So if I’m a bit stiff in the arm or I’m moving it weird, it’s just trying to stretch it because it gets a bit sore.

Another cool fact about me is I used to work 10 years in the hospitality industry. So I know a lot about different beers and that’s where I really got my marketing from. So everything I do now is taken root from that time in hospitality industry where I was running beer festivals and I was running a pub. So yeah, it’s a little passion project there.

Frederick Vallaeys: Interesting. Well, the UK is not a bad place to live if you like beer, right? A lot of beer in the blood.

Anyway, so let’s bring in the slides and I’ll give you the stage and tell us a little bit about—so we’ve covered who you are—let’s talk a little bit about the automation layer that you’ve built.

Chris Ridley: Yeah, absolutely. So we built an automation tool called the Search Cannibalization Management Tool. And this all came about when one of our clients, Puck Stop, wanted to run Google Ads but they already had a really strong organic search presence and they were afraid of Google Ads stealing their traffic that they’ve worked hard on—that our SEO team at Evoluted has worked hard on—and we wanted to reassure them through automation that we won’t let this happen.

As head of paid media at Evoluted, I got to work with our development team at Evoluted and our organic search team. We had a partnership with Puck Stop for multiple years at this point already. And the product I’m going to showcase today is the creation we made from this collaboration of all three of us working together to make sure that we protect our clients’ budget and we get incremental growth from the investment they’re putting into us.

Frederick Vallaeys: Yeah. And what I want to maybe preface here is some people think you should always run SEO and PPC at the same time. Other people think you should only do it in certain scenarios. There should be certain controls. Ultimately, that’s a whole different conversation. We’re not going to have that conversation here. But I think the point to what you’re going to show is once you as an agency or as an in-house team come up with “here’s what we’d like to do,” that can become very tedious, very manual. And that’s where automations can come into play.

And so you’re going to take us through one of the ways to approach this. That’s what I love about it—find a problem, find out what we want to do about it, and then let’s try to make it as easy as possible to actually deliver to the client.

Chris Ridley: Yeah, absolutely. So Puck Stop beforehand never ran any Google Ads. And some people say you should always run Google Ads with your organic search. You know, we always have clients that run both, but we know there are certain instances when you don’t want both to be running at the same time. One of them is when you’ve got really good organic presence for a certain search theme or topic and very little competition via Google Ads, or it doesn’t matter the competition because of the brand term or the topic you’re doing and you’re seen as a leader in that market. People are going to click on yours regardless of ads shown there or not.

So while this is automation, there will be some manual aspects where you can fine-tune and tweak when it appears and the conditions to it as well.

Frederick Vallaeys: Right. Human in the loop. Puck Stop was the client in this case. So because you’re going to be referencing them quite a bit, tell people what Puck Stop does exactly.

Chris Ridley: Yeah, I’ve got slides here. So Puck Stop, just in case anyone’s not familiar with them, are an online and brick-and-mortar ice hockey specialist equipment store based in Sheffield, just around the corner from the Sheffield Steelers, the local ice hockey team. And they recently migrated to a Shopify site before we did this project, which was the perfect opportunity for us to test this new solution and really push it.

They sell stuff from ice hockey skates to pucks to protective gear, gloves, merchandise and sticks with average order values from—you know, could be just £20-£30 all the way to a few hundred pounds as well. So that gives a bit of context when we talk about ROAS and average order values as well.

Frederick Vallaeys: Great. Okay. So let’s go back to the big problem that you discovered or that you were trying to solve with them.

Chris Ridley: Absolutely. So the big problem we faced with Puck Stop was first a bit of hesitation in investing in Google Ads after already having such good strong organic ranking performance and they didn’t want any cannibalization to happen. Once we reassured them that this solution would help, we had to work out a way to make it work—the actual cogs in the background.

And the best way we found to do this was through an API with our SEO keyword tracking tool, SEO Monitor. By hooking up their free API to a Google Sheet, we were able to take a daily picture snapshot of our organic rankings. And there we noticed a pattern where from looking at the SERPs as well, if their organic ranking dropped below two or three, they’re actually below the fold. We noticed their clicks were dropping off a lot at that point. And we noticed a lot of these also had their paid listings at the top.

And this is where we came up with a solution to start using organic rankings to determine when our ads would turn on and off and whether we wanted to turn search ads and shopping ads off or just search. So this works with both e-commerce and lead gen. You can make it impact and work with your shopping ads or just your search ads.

Frederick Vallaeys: You know, I’m curious here. So when you think about segmentation in PPC, you can put in audiences. You can obviously put in geo-targeting and you can get fairly specific and run different campaigns for different cities, states, countries. How does that generally hook up with an SEO tool? Does that mean you have to replicate to those same segments basically so that when they go and evaluate “are we top three?” it is looking for a specific audience, specific geographic area, or are these ranking signals more generic and then they get deployed across many segments where it’s not 100% certain that for that segment that was the exact scenario?

Chris Ridley: A really good question there. And through SEO Monitor, you can set up multiple projects which track keywords and you can base it on cities or other locations. So for Puck Stop, as they have a limited budget, they don’t do a lot of segmentation in terms of targeting by location just because their budget wouldn’t go far enough. But what we can do in SEO Monitor—bigger clients, clients that want to make sure that they’re not turning keywords off in their most competitive areas, say London or other big cities—you can set up an SEO Monitor project which is targeting, say, London’s organic ranking, and you could pull that through with a small tweak to the keyword label that we applied to the paid keyword.

You could then make sure that it only gets triggered by that JavaScript. So you could have multiple of these solutions running, one for Birmingham, one for Manchester, one for London, or do it based on international cities or international countries. And you can have it running multiple—one of these layering on top of each other so that each geographic area market that you want to have a different strategy to, you can do that.

Frederick Vallaeys: Makes sense. Okay. So let’s jump into that next layer. You’ve evaluated “is my keyword in the top three or not?” You mentioned something about putting labels on. So I think that’s going to be another layer, but what’s next after you’ve done this step?

Chris Ridley: Yeah. So the way that we approach it is we’ve built a Google Sheets template which has the JavaScript already built into it. A link to that will be in the show notes of this episode. And we also give you a guide as well on how to set up the SEO Monitor API. It’s fairly simple, just a quick Chrome extension to download, and then you can go from there. And I’ll talk through the setup as well within documentation that will be in the show notes as well.

Once you’ve got that, you should see—at this point we’ll want to jump to my screen share—you will see your SERP cannibalization being pulled through into a spreadsheet. So once you’ve set up the SEO Monitor API within your Google Sheet, you’ll see something like this, where it will show you the search terms and keywords for the last 29 days, including your mobile ranking data, the opportunity (so how hard it is) and all that, which we don’t need to worry about, and any groupings you’ve done.

You can also use these groupings to help you segment your keyword based on themes, ad groups, campaigns as well. The key ones we’re looking at at the moment though will be Column A, which is keywords, and then we’ll be using mobile rank as the spreadsheet pulls these out separately here so you can see what they are. My tip here is always to sort it by lowest to highest. And then we can then go through here and applying labels to these, which are these.

So these are once you’re on here you’ll have all of those organic keywords here and labels will be automatically created for them in here. The part you’ll have to manually do is go through your keywords in your Google Ads account and apply these to them. The best way to do that is using Google Ads Editor, a free tool that lets you make bulk changes offline. So I would do it at ad group or campaign at a time. Try it on a small level and then scale it when you see the success.

So for this example, you might want to look at the bespoke PHP development ad group, apply all the labels within that ad group, and then push that live into Google Ads.

Frederick Vallaeys: Okay. So let me ask you a few questions. First of all, you mentioned there’s an API integration into a Google Sheet. So for people who haven’t done that, is that pretty complicated or do you just need an SEO Monitor account, which is a paid account?

Chris Ridley: We are looking at integrating SEMrush and Ahrefs as well. I’ve reached out to them. I’m just waiting for their tech team to get back to me about how they want to implement it.

If you’ve got other keyword ranking tools out there that you want integrated as well, drop Fred or myself an email and then we can get work on getting that integrated as well.

Frederick Vallaeys: Yeah, exactly. And so I mean it’s just kind of like a function, right? Where you would do an easy function like SUM. There are some API connector functions and I believe that’s usually how you—

Chris Ridley: So with this one it’s just an extension you get for Google Sheets. You then go into API authorization. In here it will tell you to go to your profile…

Frederick Vallaeys: Did you build this extension or is this from SEO Monitor?

Chris Ridley: This is SEO Monitor, so they’ve got a copy of the playbook, of the guide that I’ve written on their website. This is theirs. You go in here, you copy your API token. You then go back into here, paste it in there, and then you go back into SEO Monitor, go create new report.

Frederick Vallaeys: Yeah, that was going to be my next question then, right? So the keywords that you see on here, that is driven by what you’re tracking in SEO Monitor. So if you only wanted to track certain keywords, the SEO tool handles that. This is just a representation then of whatever’s in that tool.

Chris Ridley: Exactly. The API just pulls the data that you’re tracking from SEO Monitor into Google Sheets. So you’ll have to set up the keywords you want to track in SEO Monitor first. Same like you would do with any other tools like Ahrefs, SEMrush.

And then what you do is go into the extension, go create report for the first time, and then in here you put the name in, and that will impact what name appears on the tab. You then select the company you want. So in this Evoluted, then takes a second because it loads the data. You then choose which business you want to choose. So let’s go Evoluted here. Choose keyword performance last 30 days. You can determine if you want to do different groups or if you just want to do all the keywords and then any filters.

So say you don’t want it to impact your brand. You can just then put a filter in here to exclude Evoluted in this example. You then make sure this is daily. You press update and it creates this sheet for you. You don’t need to edit anything else in these cells and these tabs. Does it all for you through the API.

Frederick Vallaeys: Nice. Okay. So we’ve got that data coming from the SEO tool and obviously if you use a different SEO tool, the logistics might be slightly different. But a lot of these concepts are pretty consistent, right? So should work for many platforms. Now, that next tab that you showed us on the spreadsheet had to do with a label that you put in. Is that label automatically generated or what’s going to be the purpose of that label? How do we set that up?

Chris Ridley: So within this template, these labels will be automatically generated as a concatenate function. It simply pulls through your keyword from the other spreadsheet as well as the mobile ranking and then it just concatenates a prefix which is “SERP-” at the beginning. The reason we use this prefix is within our script, which you can find on the README tab, and it’s all color-coded to help you amend it and tailor it to your need.

You’ll see in here we have a label checker. So it’s only going to impact, trigger on keywords that have labels that start with “SERP-”. If you are running multiple versions of this, say you want to run one for London, you want to run one for Paris, you want to run one for Munich, you can change this to be—let me open it for you—could go “SERP Munich” and then in here if you just change all of these to be “SERP Munich” to match this can then be your Munich keyword tracker as long as you pull through in your report just the keyword organic rankings in Munich.

You could then set one up for London, Paris, Madrid, Barcelona, wherever you want. You just need to do a different JavaScript each one changing what it says here.

Frederick Vallaeys: And to maybe up level for a second, so I think what you’re doing here is you’ve got your Google Ads data and you’re trying to bring in the SEO ranking into Google Ads. And so one way to bring external data inside of Google Ads so you don’t have to connect two systems later on is to use labels. And so you’re setting a label name and then for each of these label names, you’re going to periodically look at what is the ranking from SEO Monitor and then you’re going to put that in as the value for the label. So that effectively brings that piece of data into Google Ads so that you can now start automating something with that. Did I get that right?

Chris Ridley: We don’t bring the value of the rank into it. We bring the value of the—we bring the keyword name into the label. What we do is we pull organic ranking into Google Sheets and we then set up a JavaScript in Google Ads to check what the organic ranking is for each of those organic keywords. We then use labels to tell Google Ads which keywords should be paused if one of these are above the threshold we set.

So in here, if we set this as two and that is three, what we’re saying is if a keyword has organic rank one or two, pause the keyword. If it has an organic rank of three or lower down the page, four, five, six, seven, turn that Google Ads keyword back on. But we only tell it to do it to keywords that have a label with the prefix “SERP” and only if the label matches the label here as well in the rank.

So for this one it would be paused. For this one it wouldn’t get paused. So if I applied this to a keyword that was say “tech SEO consultancy,” this would get paused. But if I had one that was “paid media consultants,” it wouldn’t get paused because that’s rank four.

Frederick Vallaeys: Yeah. Okay, great. Thank you. Thanks for clarifying that. Okay, cool. So we got this spreadsheet. So what’s next?

Chris Ridley: So once you’ve got the spreadsheet set up, you then edit this Google Ads script, which is all formatted for you. You just need to make a few changes here. Just the red bits here and these. This blue value has to match this blue tab as well. So this tab has a little blue underline to help you understand that. So if you change the name of this, just make sure you update it in here.

The only other thing you then need to do is go into Google Ads, put this into the script section within Google Ads, test it, make sure it runs, and then apply keywords that you want to target with labels. So like I recommend, I recommend doing this on a small scale first. Make sure it works on a small scale and it’s not letting you lose a lot of traffic. And at that point, you can go into Google Ads, apply those labels, upload it, and test it.

Frederick Vallaeys: Okay? And so that’s what this visual here represents. So we’re pulling SEO data into the Google Sheet, and that has a two-way connection into Google Ads to maintain the decision about pausing or unpausing depending on what data is in that Google Sheet from SEO Monitor.

Chris Ridley: Exactly. If it will help I can show Google Ads Editor as well and show how I would copy and paste that data from Google Ads Editor by labels. So I’m sharing the screen now which is Google Ads Editor. To access this all you need to do is make sure you download the tool. It will then ask you to log into your Google Ads just like you normally will. It will ask you to download the account. So make an offline version of them. For this I’ve taken Evoluted. You want to make sure you get the recent changes which I’ve done.

And what we’ll do is we’ll just choose a small selection of keywords like these. Copy. For this you can just make a new spreadsheet and paste them into here. And then what you’ll want to do is have the label value here. And you’ll want to go through these labels and find the one that is most relevant to it. And this is why it’s best doing it by ad group because you’ll have a theme there. So with this one, these are web development. And I can just go through these and even apply a filter here. Go in development.

And then I can go through seeing which one I think is most relevant. Another good way of doing it as well is just checking what comes up when you check the SERP. So you put it in, go through, you can see they’re talking about agencies, services, agencies organically. It is company and agencies. So any of those might be worth looking at.

Frederick Vallaeys: So this is interesting too because you’re basically not doing a keyword rank or sorry an SEO rank for every single keyword because that might just become too much data. So it seems like you’re finding the theme for your many, many Google Ads keywords and then categorizing them here. And so it looks like you’re doing this manually.

Chris Ridley: Yeah. The reason we’re doing manual here is we’ve tried to use LLMs and other automated ways of matching language and we found that it can get it wrong. So for Puck Stop they do ice hockey only. So something like “hockey skates” you know when you search that you most likely get ice hockey skates coming up. However, an LLM or generative AI might assume it is like roller skates and it’s like field hockey in that way which can then make the mistake because it doesn’t understand the context.

So we found the best way to make sure you are targeting the correct SERPs is by doing this bit manually. If you have a huge account, you know, it is a bit of a pain, but we always recommend start it small, check it isn’t having a negative impact on your performance and scale it up.

We are looking at ways that we might be able to automate this. It’s just at the moment technology is not there for us at the moment where we feel confident that it’s going to match it how we want. The other thing you’ll have as well is your SEO team won’t be tracking keywords for every different search term query and every variant of a query that comes up. But a lot of them fall into the same basket when it comes to organic ranking. And your keywords in Google Ads don’t ever really match what you put into your SEO ranking tracking tools or what your client will be asking about. So there’s never a good marriage there.

I did show a Spider-Man meme where it was two organic keywords both pointing at the same screen to take a look at.

We got “hockey skates.” Exactly. So you can put “hockey skates” into Google and you can put “hockey ice skates” into Google and they both come up with near enough of the same result. However, which one would you apply the keyword to for your Google Ads? Because they’re both relevant. The best thing to do is choose the one most relevant. And that’s where you have to work down and work out what’s best. You can’t leave AI to work that out because it doesn’t understand the context around the difference between ice skates and skates when they both got hockey.

Just a side note, we even tried tools where you try to match the words up and even that was getting tripped over because if the words are in the wrong order that the machine was just not understanding it.

Frederick Vallaeys: And that’s a bit of challenge with all the fuzziness around match types. Because now what you’re doing is you’re pausing a certain keyword that corresponds to an SEO rank of the same keyword or a similar theme based on what you as a human decided but then you get overlap right and I think you had some slides about Performance Max but even on a broad match it may just start pulling in a different keyword that’s related. So how do you manage those kinds of elements?

Chris Ridley: Yeah. So ultimately because of all the overlaps because the different keyword match types as well and now we’ve got AI Max for search coming in which is just adding the overlap even more. The way that we approach it is understand what the SERPs are that show up when people are searching for your Google Ads keywords and try to find the best organic keyword you’re tracking to marry it with.

When it comes to Performance Max, this is really hard because there’s no keywords in there for you to apply labels to. There’s no keywords there for you to control. So our solution there is by building an automated negative keyword list for your Performance Max campaign. If you’re unable to build that automated negative keyword list for Performance Max, maybe you don’t have access to that feature yet, maybe you don’t have an ads rep to do it for you, you can resort to an account level negative keyword. That would impact your shopping ads though as well.

But these are the ways we’re doing it. In terms of the automation, you can use the exact same tool, but tweak it slightly, which we’ll release later this month, where it will create an automated negative keyword for you based on your organic ranking. And you do the same thing if it’s ranking one or two, you make it a negative keyword. It’s ranking three, four or lower, you remove it from that negative keyword list. So we’re going to be working on that tool and releasing that publicly as well as soon as we can.

Frederick Vallaeys: All right. So I think we understand how these pieces fit together. And by the way, what’s the piece that does the pausing and enabling? That’s the JavaScript.

Chris Ridley: Yeah, the JavaScript.

Frederick Vallaeys: Great. And so you put that on what sort of frequency? I guess once an hour is too often, maybe once a day. Is that what you recommend?

Chris Ridley: Yeah. Yeah, we set up the Google Ads JavaScript to run once a day. I set it for 7:00 to 8:00 AM because the SEO Monitor API pulls are around 6:00 AM. So if you want to really play with fire, you can try to set it 6:00 to 7:00 and hope that the API pulls before the Google ad does. But I find setting at 7:00 gives you peace of mind that it’s running on the latest time.

You can also check when your API has been updated because in the top right corner on the tab, it will come up saying “last updated August 8th” or “last updated September 7th.”

Frederick Vallaeys: Right. So that’s here in the spreadsheet. It shows you right there when the last update pulled in.

Chris Ridley: And it also lets you know when it breaks. So your API token might expire. You might lose access and you might have to change a login. So if this doesn’t update and it’s showing an old date here where it says “next update,” you know it hasn’t been refreshed. A way you can fix that is by going into extensions, going to SEO one and refresh current report. Then it will throw up any errors if it’s not working and tell you what’s wrong.

Frederick Vallaeys: Cool. Great. Thanks for showing us that. Okay. So you did all this work for Puck Stop and million-dollar question now. Did it work?

Chris Ridley: Yeah, really good question to ask. And it did work. We actually won three awards at the Northern Digital Awards event in 2024 around December time, including silver for Best Digital Marketing Campaign Retail, gold for Best Marketing Campaign B2C, and we also got Best Integrated Campaign because it involved dev, SEO and PPC.

We also got some great feedback from the judges as well at the event who were praising us for the real-time data usage, the innovative SEO and PPC integration and how we used API to overcome creative problem solving and client hesitation over introducing Google Ads to an organically strong website.

Frederick Vallaeys: Congratulations. Well done. And then what I love about it is it’s doing two things, right? So it is automating a process—you know it is still keeping the human in the loop to make some of the decisions and then the fact that it’s really bringing in your own data. So first party data, a lot of companies think about that as like maybe their Merchant Center, maybe their CRM lists, but your SEO data is also first-party data, right?

So if you can connect that into Google Ads, that is yet one more way that you can set up a little bit of an advantage against everyone who’s not necessarily doing this and using that factor as a decision point in how they run PPC ads. So really love it for all those reasons.

Chris Ridley: Yeah, absolutely. And it is something that we’ve been doing at Evoluted a lot. When a client tells us that we’ve got really good organic presence or they don’t want to bid on brand terms anymore, we’ve manually turned them off, only then for the client to say that an update’s happened or the SEO team to tell us an update happened, rankings have dipped, and then you have to reactively turn them back on. You’ve got that lag between when the update happens and when you actually implement it.

There will no longer be a lag between that because if your organic ranking drops because of Google algorithm update, because of the change to the web page, web page going down, whatever like that, straight away it’ll kick in. You know, the next day within 24 hours, it will kick in and you already have it there. And like you said, this is first party data a lot of people don’t consider. So why not use it?

And it can go beyond just determining if you want to stop cannibalization on brand terms. You could do it for other stuff as well. So for us, we got the target of increasing total search revenue across the entire website by 25% year-on-year. And with the small migration period you have, you always have a little wobble with your organic rankings as Google gets familiar with the new website. And our other objective there was recover that. The big part of that was don’t let Google Ads steal that traffic and don’t let Google Ads steal the performance that we were getting organically.

And we actually saw 113% growth in search revenue. And the budget we put behind this wasn’t big. It was only a couple of thousand pounds in media spend. You know, we’re not talking tens or hundreds of thousands per month. And because of the great average order value we could get from the business as well, because of the ice hockey and the skates and all that, we got a really healthy 27x return on ad spend, which was a lot more than the client expected, even with those big average order values.

And overall, because of the changes we made by going to Shopify, small changes the SEO team made on the website as well, we saw conversion rates increase for the entire website by 11.66% as well. So it wasn’t just a Google Ads success story. This is a success story across all of our departments working with Puck Stop. And it resulted with the client being really happy and us making a road map for where we want to take this tool in the future. So it isn’t just useful now, but it is useful in the future, no matter what industry you’re working in or what budget size you have.

Frederick Vallaeys: And so yeah, you’re showing here on the screen what that road map looks like. One thing that I’m curious about is if you’ve had a conversation around generative AI search results and AI mode, which for some countries is really just rolling out. So it’s got a little ways to go before it takes over. But yeah, what happens if AI mode turns on and it’s not really about being a top three organic result anymore, but it’s about are you visible in that AI response. Have you thought about how that might be integrated?

Chris Ridley: Yeah. So we had a really good conversation with Jack at Candor about this on a recent podcast as well where he asked, you know, with generative AI now and these AI overviews, ranks don’t matter. It’s about getting into those sections, getting into the actual conversational SERPs.

At the moment in Google, the AI mode is a separate tab to the search mode at the moment. And when you go on to Google and you search, it defaults you to the search tab still for now. I imagine this will change where it will default you to the AI mode. There’ll be a bit of backlash from everyone, publishers, SEO teams, everyone and they might switch back.

Frederick Vallaeys: So when you just said everyone, but it was what the publishers and it was like stakeholders. What about the consumers? I think the consumers might like it.

Chris Ridley: Yeah. So I think what it is, is the consumers might grumble about it and I might be wrong here. I think they’ll grumble but they aren’t listened to as much as the SEO people and the Google Ads people because we have people like Ginny Marvin who is an ads liaison. We have John Mueller that we can get in touch with. A lot of people don’t know who these are. They don’t know that they can contact Google. It is seen as a multi-billion pound international company. And maybe this is a cultural thing, but in the UK, we prefer to grumble a bit more than to go to the top about it. We’re too polite for that.

So I see change happening when those people using the tools and those people making content for Google when they complain that’s when I think Google contacts notice just because there’s not the channels out there at the moment for consumers to raise their voice as well.

Frederick Vallaeys: Right. And I think the consumer voice to some degree at least when I was at Google was captured through looking at the metrics. Right? What is the user behavior change when we move these pixels around or when we make this tab a default or when we make that color change? And so obviously I think they’re going to be monitoring very closely what’s happening in terms of search growth with Perplexity, with GPT, and do anything they can to keep Google relevant as well and so AI mode is a good alternative to those other two that I just mentioned but they are still very conservative in that they hide it behind the tab and you have to make a decision to go there and then there’s the AI overview which does show on the main tab.

And that also to some degree detracts a little bit from the ads right? Most people who started engaging with the AI overviews but also not really look down the page for those additional ads and SEO links. So it’s an interesting and evolving space and maybe Evoluted—your company name is a good one because there’s so much that keeps happening.

And ultimately I think you keep on top of what’s happening. You have a tool which—thank you for open sourcing it. So that means anyone can grab it. And if you have an insight and you think you can do something better with it, you can do that now, right? And it’d be great if you brought that back to the community as well. Anyone listening making those changes? But yeah, it’s going to be interesting with all of these AI overview and AI mode things happening.

Chris Ridley: Yeah, I see AI overview is just an evolution of the zero clicks results we’re seeing anyway from all the featured snippets and everything. They’ve just expanded on that and being more dynamic. And I wouldn’t be surprised. They’ve now got it on a separate tab. I wouldn’t be surprised if Google—and you’ll know their thought process and their logic a bit more because of your experience there—I wouldn’t be surprised if they evolve AI overview to have a call to action which is, you know, “learn more” or “discuss it with AI” with Google AI type thing. Maybe they will have a thing there. So rather than you doing a new search, it just takes you to the AI version and you can actually have a conversation with it about the topic you just searched.

That would save them having to have that separate tab hidden away. And I think that could be a good way for Google to turn a search into a conversation.

Frederick Vallaeys: Yeah. Well, and my main frustration right now in AI overviews is that it doesn’t have the history. So it’s not truly a chat. It is a generative summarization of what might be a useful answer, which is helpful. But the whole nice thing about conversations is you can go deeper. You can ask a follow-up question. And that is missing in AI overviews. And that’s why I think AI mode is probably going to be the way that we head.

But it’s also one of the reasons why the individual keyword is becoming less telling because ultimately the ads that Google should show, the organic results they should show are influenced by that 10-prompt conversation that you just had and everything that happened in that and maybe even a conversation you had last week about a similar topic. So it just becomes a little bit harder as an advertiser to say like “okay this successful sale can be tied back to this specific keyword” when all of that becomes a bit fuzzy as we shift to prompting as opposed to keyword searching.

Chris Ridley: Yeah, I remember five, six years ago where home devices, you know, Amazon Alexa, Google Home and all that became popular and every agency was talking about how voice chat was going to replace Google and how it was going to destroy SEO and how it was going to really make Google Ads hard because you were going to have a big question with 20 words now and would it work and it never happened you know and I think now this is happening people are actually having verbal conversations with AI. You know, they’re showing it in the ads as well when they’re showing the Google AI and they’re showing complexity where people have conversations.

And I think we’re—I think Google Ads is already priming itself for that by doing topics and not allowing you to choose keywords. I think audience targeting will become a bigger thing as well. And they might need to expand their internal audience list that they provide in terms of like interests and affiliations and all that. That would be a big thing if we can start actually seeing what people have regular conversations about. If someone keeps talking about the same football team, you know, that’s when you want to start actually prompting them about going to games to it or, you know, buying merchandise or fan-made stuff for the team.

And maybe if they keep talking about visiting a certain city, you can then prompt them about, well, look, it’s actually price drop on hotels in the area or price drops on flights. So those could be really interesting as well.

Frederick Vallaeys: Yeah. Yeah. And then I suppose what’s nice about the solution that you’ve built, it really does work for keywords today. But if you say wanted to apply this to more of a generative search mode or answer engine behavior, then you just swap out that front end piece, which is the one that looks at what is my SEO ranking in the traditional historical SEO rankings that we’re used to. You swap that out for some newfangled tool that tells you how well you’re doing on AI mode.

And that might become your toggle then that says “okay well in that case we should advertise, in that case we shouldn’t advertise.” So it’s very modular and again because you’ve open sourced it it’s just a matter of figuring out what is the function that I need to change and by the way you don’t even have to do it yourself you can go and have generative AI—I’ve given it scripts and I’ve said “here’s a script I have can you please update it so that it does blah blah blah” and I tell it what needs to be changed and it writes the code and it generally works so that’s another way to do it.

Chris Ridley: Yeah. So I’ve actually used it a little bit when I’ve had to tweak the code as well. So I noticed that it would get into a bit of a loop where it was going through every label we had in the account. That’s when I had to add the prefix. I then made sure that I just sense checked it within ChatGPT. Made sure that the change I made to the code was nice and simple and it didn’t break anything just because it was something I wanted to be able to do quick because someone raised it to me and I didn’t want to have to book it in with the devs and wait for that.

Good thing about it as well is because it’s used in labels, you could change it from a keyword label to an ad group label, you could change it to campaign label. So even if keyword come obsolete in Google Ads, you can then just take this to a different asset that you can apply labels to. Like you said, it’s completely modular.

Frederick Vallaeys: Yeah. Future proof. All right. Well, hey Chris, fantastic stuff. I really appreciate you putting together some slides to show us the concepts, but also showing us in the interfaces itself and the different tools how this all works. So all of the resources to be able to do this yourself are in the show notes. So grab those links. You can also connect with Chris, head of paid media at Evoluted through that QR code and his information is also in the show notes. Chris, any final thoughts before we wrap up here?

Chris Ridley: Just if you get a chance to use the tool, let me know your feedback, whether that’s through LinkedIn, whether that is through email. Always love to hear how people are tweaking it, evolving it themselves, or if there’s anything you want help in adding, I’m more than happy to jump on a video call and work through it with you as well.

Frederick Vallaeys: Great. Very generous. Okay. Well, everyone, thank you for watching this episode of Automation Layering Masterclass. If you thought this was useful, we’d love to get a thumbs up and also subscribe so that you’ll see future episodes of this. You can also go on the Optmyzr website and all of the episodes are listed there. I also have the PPC Town Hall podcast that’s bi-weekly. So go ahead and look for that one if you want more of this type of content. With that, thank you Chris. Thank you everyone for watching and we’ll see you for the next one. Cheers.

Chris Ridley: Thanks, Fred.