From SEO to GEO: Winning in the Age of AI Search with John Campbell

[00:00:00] John: SEO manager would brief in that copywriter. Now they're briefing in that tool and we've reduced the cost 'cause it's overall being cheaper, but also that used to be a back and forth of an email to that person. They'd have to wait for that person. Sometimes they'd be abroad and stuff like that. So it's brought.
Down the time to getting stuff live and f first for our clients. The faster we can get stuff live on their website, the sooner we'll see SEA results, et cetera.
[00:00:27] Mike: My name's Mike Lander and you are listening to Higgle, the B2B Sales Club podcast. Where we bring you actionable insights about sales, RFPs, negotiations, and difficult procurement discussions from sales leaders, ground leaders, and procurement leaders.
Please subscribe to get updates when new episodes are released.
John, thanks Everance for joining me on haggle the B2B Sales Club podcast. Great to have you.
[00:00:55] John: My pleasure to be here. Thanks for the invite.
[00:00:57] Mike: Pleasure. So fascinating topic today, obviously all about AI and search, but before we do that, uh, who are you, what do you do, and what's your favorite song and why?
[00:01:06] John: Oh, three good questions.
So my name's John Campbell, head of AI and Innovation at a company called Roast, a performance marketing agency based in London, Cape Town and New York. And so my role is all around how AI is changing the way that. People are searching. And so that infect our clients, but then also to see how that shapes what we do internally in terms of our roles, the services that we offer.
Where do we use ai? Where do we not use ai? How does that make us more efficient? And how has that implemented across a growing agency group? So we're getting more and more offices, and as that happens, that'll be even more important.
[00:01:45] Mike: Brilliant. Perfect. And that's gonna be a really, really interesting discussion 'cause I'm having lots of thoughts, reading lots about AI written papers.
So yeah, really interesting about that agency landscape and where it's going.
[00:01:56] John: And then last thing, the song, I think I'll go for Fat Boy Slim right here, right now. Oh,
[00:02:01] Mike: brilliant song.
[00:02:02] John: A good song to kind of get you pumped up. And um, back in the day when I was younger, I used to be the Match day DJ at Blackpool Football Club.
So like
[00:02:13] Mike: poor football? No.
[00:02:14] John: Yes. So a good friend of ours was the. The announcer and then, yeah, when we moved into the new stadium, he was like, John, I don't have a clue how to use any of this new technology. Used to have a tape player or something. Yeah, yeah. So I used to sit up there, got my tickets for free, and that was the song that we used to play as the players would come out.
So I used to get new comes up, but it still does now.
[00:02:33] Mike: What a great story. I was born in Blackpool.
[00:02:35] John: Oh, there we go. Me too as well.
[00:02:37] Mike: There we go. Were you?
[00:02:38] John: Yeah. The great toilet hospital is on the passport. Yeah.
[00:02:42] Mike: Wow. So I was born in Blackpool, then I was adopted at six months old and went to Manchester. So I have no kind of like history of Blackpool at all.
Mm-hmm. But yeah, I was born there. So there we go.
[00:02:52] John: Yeah. Great place to come from.
[00:02:53] Mike: Great place to come from. Precisely. So let's get into the questions. Just talk about the kinda history of search. You know, where it's come from. Where's search going? Especially now with AI powered search and kind of just like, let's bust some myths if we can explain GEO organic search, all that kind of stuff.
[00:03:12] John: Yeah. So I think a good paradigm that we're kind of, we often talk about is. Imagine if you go to the library and you're looking to kind of research on a particular topic. Maybe it's um, you know, world War II or the war in Vietnam or any, any kind of topical subject. And if you think what would happen in the past, you'd go in the librarian, maybe here's 10 bucks.
You go read them, you go and review 'em. And that's kind of what Google was in the past is Google was there to say, okay, we're gonna retrieve. 10 blue links, 10 results, we're gonna give it to you and therefore, you know, it's up to you. Then to decipher that and read it and work out what's important. We've tried to give you the most important results, but now what we've seen is come is to like a pa, so personal assistance where you've asked a question, the PA then goes, get those 10 books, reads them, reviews them.
Puts the bits that are most important, puts into a nice bullet point list, summarizes it all for you. So we've gone from this paradigm shift in the way that we retrieve information, and as a result of that, the old model was that you, as the website owner, the benefit was that you've got the click. The visitor came through to your website, found out about you retargeting.
Whereas now we've kind of gone into this system where you might not get the click and you're not getting the traffic much. So zero click results exactly. Zero click results. And so therefore. It's affecting a lot of companies that organic traffic is down, they're seeing less click through rates, but at the same time, you know, the quality of the traffic is probably getting better because you had a lot of people who were just clicking, looking, not the right thing.
Whereas now the people that do come through that journey and click through, they're a little bit more qualified, and so there's a huge amount of interest in how do you get the most of that traffic. It's still early days. But then you've got this whole problem of, I used to, you know, judge my organic traffic performance on organic non-brand clicks, increasing year on year and year on year.
And you know, I want to have some of my traffic, which is brand traffic, some of my traffic, which is people have never heard about me coming through. And that system's kind of slowly changing. And you know, obviously that's a big thing for everybody to kind of get their heads through at this moment in time.
And. It's hitting different industries and different verticals in different levels. Um, so you know, if you're high-end fashion, maybe it's not impacting you too much 'cause you might not go to chat GTP for fashion recommendations. But then if you are buying a complex kind of B2B products and you are looking at, you know, what different, comparing different providers, seeing how that pro provider's solution fits to your particular business, it's a really good fit.
For, for search. And that's why a lot of people u are moving to AI to find out about products and services.
[00:05:48] Mike: Ah, so let's go back a stage. Let's look at the buyer journey. I was at a conference last week, run by Launch Online, who I work with all about, basically performance marketing, and there was lots of talk about.
The buyer journey used to be that kinda light awareness, interest, desire, and action, whichever model you chose. But it was quite a linear journey. And now they're saying, oh, it's not quite like that because people are doing more discovery based search. You know, they're doing more curiosity based search, and then eventually they're gonna buy something when they're ready to buy.
So the search characteristics don't have economic intent behind them. Early on, but then they're gonna wiggle around, try different things, look at TikTok, look at Facebook, do all sorts of things. Is that what's going on? And how does AI and kinda like CHATT and Claude and Gemini fit into all of that mess?
[00:06:38] John: Yeah, we see some of that definitely happening. People are aware of the businesses and they may discover them earlier on, you know, with search, but it might be earlier seen an event or they, they saw 'em here, there, and in different places. But what we are seeing towards the end of the journey is people are going, oh, right, I need to, I've got four people, or four different companies that I've shortlisted.
I normally have to. Google each one of them and then Google them reviews, then Google, like watch their YouTube videos, then Google their name with customer support. So I've got four things and I've got four companies all of sudden that's 16 different searches. So what we are seeing people doing is kind of going naming company 1, 2, 3, 4, and then asking AI to do kind of a service review of them or just say.
Versus exactly X, Y, and Z. And so people are understanding that you're basically shortening that time where you'd have to do lots and lots of Google and lots and lots of, um, pulling that information together, maybe putting it in your own spreadsheet. And now what they're seeing is the ability to do, to get AI to do that comparison.
And then it's almost the user feels like. They've been part of that journey 'cause they've decided on the short lists, but then they're using AI to get 'em to compare, you know, is it the contract lens, is it the, depends on the services, it like, oh, how many key phrases I track? Or is it the service? How many territories and or the cost per seat, whatever the product is.
And so then people are then taking that away. And then probably, you know, most buying cycles, you are not the decision maker. You've gotta go to somebody to say. Then you've almost got your workings out to say, well, actually this one's more expensive, but yeah, X, Y, and Z. This one's cheaper. I think this will do for us now.
And then maybe in New Year's time, we'll move to this next product. And so we're starting to see that in the prompt data. But as a result of that, you'll see a lot of businesses now are trying to. Produce some of that content where they are on their own website saying, US versus competitor X, Y, and Z.
[00:08:31] Mike: So this is the next question is like, what happens?
How do you, so bear in mind listeners, I know very little about search. I know a bit ish, but not a great deal. How do you therefore, if that's going on, and it's all about user driven discovery and selection. So we're no longer just doing discovery, we're now doing, let me use some intelligent tools like.
One of the foundation models to help me down, select to basically the one that I really want, and then eventually I'm gonna go onto a site and actually buy it. So how can businesses, should businesses, how do they A, make sure that you know, one of the LLMs, all of them. Have them in their index, and then two, how do they then encourage the LLMs to pick them as the right version?
How does all that work?
[00:09:21] John: So fundamentally at the moment, the way that the majority is these models work is that they have this reasoning model, but in the background, when you ask them a question. They're going away and searching Google and Bing in the background. So if you've got good SEO metrics, they normally correlate to very good metrics in geo or generated engine optimization.
So where we see our clients ranking first page and within the top three or four results, they'll often be the most cited links as in, so in chat UTP gives an answer. There'll be some links there.
[00:09:53] Mike: So that's 'cause I guess. As a human searching, I'd be on page one, the top five results. So it's doing something kind of similar.
It's looking for signals, what's trustworthy, what looks like it's reasonably high quality.
[00:10:08] John: Yes. And then what you'll start to notice, it depends on what you're saying, but then. You'll start to see third party websites, which are, you know, like G two or these review websites. In our industry, it's a website called Clutch, which is where, you know, they'll have all the statistics about your agency or about your product, or it could be, you know, the model looks for award winnings.
Then it's about, well, how do I win awards and then how do I publish 'em content. Fundamentally, what you wanna do is first of all, have good information about your product that's accessible on your website. So, you know, if it's a SaaS product, it's about having, you know, your really detailed, well laid out HTML table that compares the different tiers of what's in your products, et cetera.
If it's a more technical product, it's maybe let's not hide all the, the information in PDFs. Let's put that onto web pages, which is easily accessible. So I think you would always start off with your own website. That's the best asset to optimize 'cause you control it. And then the second one is. Looking at when the model is recommended and it's citing pages, what are those pages?
Is it a industry magazine? Is it a comparison website? Is it a direct to website? And then it's about getting your pages on there. What we have seen earlier on though, and as I mentioned it before, is some companies are starting to produce like list articles on their own websites. So say for example, it's like what's the best podcast recording technology?
Riverside might write a, an article and then they'll put themselves at one, and then underneath they've got Zoom. And then the other one's, the other ones, it's a little bit low funnel. 'cause it's a bit weird to be doing that on your own website. So how long that will work sustainably? It's questionable, but.
It's more about actually doing good stuff on your own website and then looking at those external websites and getting listed on there and making sure that, you know, these websites still list the pros, the cons, making sure it's all up to date and
[00:12:01] Mike: is writing content for, I mean, I've got, obviously I've got my own business.
I've got a micro website is writing content. For my website different if I'm optimizing 4G EO than it is for SEO Or should you just like, just write good content?
[00:12:17] John: Yeah, I think write good content is, is is generally the, the, uh, thing that we, we suggest doing. The only thing is you might want to think about sometimes where.
And we've done this recently with client of ours, right? Move where in the past they'd have like a monthly the mortgage rates updates and they'd have a nice graph. The graph would be an image. So now it's okay. It's underneath that graph. We have a little expanding box and it has the table. With the years, the dates and the numbers, the stuff like that.
The other thing that people had benefits of is FAQs at the bottom and then at the top, a nice summary slide. So like a summary box of content. But that's all actually when you look into it. That's useful from an, uh, from a user perspective because you know, too long didn't wanna read the whole article.
They can just read the summary. If they wanna look at the details of the graph, they can't hover over it 'cause it's an image. But they might wanna know this certain date in April, what the average price was. It's in the table. So a lot of the things go hand in hand with good content for users. Also, accessibility.
So you know, if you've got a video on your page, have the whole transcript of the video underneath it in an expanding box, et cetera. So there is some things where. It just kind of makes sense and we should have been doing them anyway. What you want to get, wanna, wanna avoid is just thinking about this from a.
I'm just writing this from a, an AI perspective. I just wanna hit the bullet points. 'cause it, it won't rank from a Google flow, is it? And it doesn't rank, then it doesn't go into the AI's consideration
[00:13:46] Mike: and you're trying to game the system at that point, which broadly in any system is a really a bad idea on the whole
[00:13:51] John: Yeah.
And I'm old enough in SEO to remember when. We went and bought thousands of links and Yes, exactly. There was a big penalty and stuff like that and
[00:13:59] Mike: he was fluffing and all that kind of stuff.
[00:14:01] John: Yeah. Yeah. It won't really help long term.
[00:14:03] Mike: It won't help in the long term. Precisely. So we'll go off piece slightly, which I hadn't prepared.
What's the impact for Google in all of this, because obviously advertising, you know, a lot of their revenue comes from advertising. It's the biggest search engine in most areas, but they now obviously have got Gemini. You get the AI results often first. Zero click, you can then interact with that I've seen now.
So I type something in, I get an AI response back. I can then interact with it without going anywhere near any adverts at all. What's gonna happen, I guess, in terms of the Google Real Estate page, the way that search engines monetize. All of that. What's your kind of view on where that's going?
[00:14:40] John: Yeah, so they've done the classic bait and switch.
So they're gonna start putting ads in the AI results. So it is a kind of that classic, here's something that's really useful from the user, but in order to sustain this and to pay for the server cost, et cetera. So yeah, that
[00:14:54] Mike: doesn't make them bad people. That's just normal commercial behavior.
[00:14:58] John: Yeah. So we're starting to see ads within the results.
Within the three
[00:15:01] Mike: versions.
[00:15:02] John: Within the three versions. So, yeah, and they started off, we all kind of like. E-commerce. So, you know, if you're asking about a cleaning product or a particular, you know, like a, an adapter or a microphone or something, you'll then get, you know, here's the AI answer is some recommendations, and then here's a selection of products, um, in that.
So they'll start to do that, whether that. Equates to the same where we had a traditional SEO search result where we'd have three ads at the top, three ads at the bottom. We interest in, uh, to see what they do from that side of things as well. And there's rumors. They've always said, do we do a paid model?
And so therefore they have a paid model where you're paying a subscription and then you don't see ads. And that's what we're actually seeing with chat GPT, is that they are gonna do an ad based model, but that's for free users, a
[00:15:50] Mike: free one. The free one is ad based. The paid one, you don't get ads. I don't get ads served on my chat GPT 'cause I use it every day for my work.
So I have a paid version. I don't get any ads,
[00:16:00] John: which has been really interesting for some of our clients because then we're thinking, right, we're Target trying to target, you know, a certain demographic. Who wouldn't have a free version of chat, GTP, but then there might be other ones where, you know, we may be targeting students or people from a lower income or people who don't really use it that much and so they, they wouldn't pay for the premium version each month.
Ads might be a good option for, for them. So it's gonna be interesting to see how, how that works. The other thing always has been with, can they sell services in other areas, like professional services? Can they sell their APIs? And subscriptions technologies. Can they sell their subscriptions for businesses like ours where we have an enterprise and account and that then they don't need to make as much money from an ad revenue, um, perspective, but most businesses will not leave money out there if there's a chance to do it.
[00:16:49] Mike: Exactly. And what impacts it having on the kind of industry at large, on the kinda on. And if we pick B2C because it's easier, the consumer industry, what impact is kind of this new world of GEO having bluntly their revenues?
[00:17:03] John: Yeah, I think it's still somewhat hard to kind of attribute this. 'cause what we're seeing is a lot of talk and a lot of interest in it.
But when you actually look at the amount of traffic that comes through to the site, it's still very, very minimal because that's somebody asking a question on an AI search and then maybe clicking through, and then from that click they go to buy. But as you mentioned before, Mike, the journey's a bit more complicated.
It might be that you discover the brand and then you open a new tab and then you search for them, and then all of a sudden that was then a maybe a PPC click or an SEO click, et cetera.
[00:17:38] Mike: That's a classic attribution problem.
[00:17:39] John: Yeah, and we still kind of got to. Work out how we attribute that. And at the moment, you'd normally do something like A MMM, so mixed modeling, but the numbers are still quite small, so then it's hard to kind of get that statistical significance to say, actually all of these visits that you had earlier on in the journey are eventually coming through as traffic, et cetera.
Um, but yeah, overall it is impacting clients because more on the, they used to get. 15, 20, 30% clickthrough rates. Position one for key phrases. And if on that key phrase, now there's a big AI overview or an AI result from Google, those click-through rates are dropping to, you know, single digits. And so that it's just less traffic.
But whether I said, as I mentioned before, maybe at the end of the day that person that really was gonna buy you ended up finding you and you're just getting rid of the people that were never a hundred percent relevant. And so there overall, it's not affecting the, the bottom line for some businesses too much.
At this moment in time. The other thing though, is if your business was fundamentally doing something that the AI can do, so we're talking to somebody in recruitment industry. They used to get a service where they'd review cvs and help you write cvs and help you prepare for interviews. Don't need that anymore.
[00:18:53] Mike: Nope. Exactly. So this is a very interesting topic. So I'm reading a brilliant book by a guy called Sangeet Choudry called Reshuffle. I dunno if you've read it at all. It's really interesting because it talks about. System level change and how enabling technologies change systems. One example is back in the day, there would be typing pools.
People would be in a typing pool and they would be typing, and they would give the documents to people to do something with when word processes came along. Now you have distributed typing capabilities, so the typing pool disappeared. Because there's no longer a need for it. So enabling technologies can change entire systems, and I think that's where the next part that the discussion is around.
A lot of people are using AI to automate what they do today. If that's client facing, that's a disaster in my view. Because what will happen is if you automate something that's client facing, you make it faster and you take away the labor model, all clients will do eventually is go by not paying you an hourly rate per person for that.
That's crazy 'cause you've automated it all using ai. So I'll have half off please. All you've done in Charlie Munger's words is you've basically, you've given all the value from the person investing in the r and d to the end customer. You've just given away all that value. So just kind of talk about your experience inside Roast of what are you doing with ai?
How do you think it changes commercial models? How do you innovate at the system level, not at the activity level. And what does that mean? I mean, it's, it's quite a big topic, John, so
[00:20:31] John: it's,
[00:20:32] Mike: but it, it's on my mind. It's fascinating.
[00:20:34] John: Yeah. So we've seen some of this happening already to the most. The easiest example is the content creation.
[00:20:42] Mike: Exactly.
[00:20:42] John: Editing. So take
[00:20:43] Mike: some examples.
[00:20:44] John: We've, you know, built a tool, it's based upon a fundamental model and then, then we've built processes and agents on top of it. But essentially what that is taking away a job of some of the copywriters. So in traditionally in SEO, it would be one of the SEO account managers say, Hey, there's all these key phrases we need to rent for them.
They put together a brief of what the page needs to talk about that would then go to a copywriter. They would write it. Edit back and forth and live. We didn't employ copywriters in-house, so it's probably an easiest decision for us. But essentially we now have a client that pays, someone did that subscription.
[00:21:19] Mike: You may have used a third party copywriting service. That company's now suffering. I'm not, this isn't a blame thing, John, at all. The opposite. I'm saying what's happening is that whole supply chain. People that were doing things before is starting to be eroded because it AI can do it instead.
[00:21:36] John: And I think what we've got is that the SEO manager would brief in that copywriter.
Now they're briefing in that tool. And two things that's happening is we've reduced the cost 'cause it's overall being cheaper, but also. That used to be a back and forth of an email to that person. They'd have to wait for that person. Sometimes they'd be abroad and stuff like that. So it's brought down the time to getting stuff live and for, for us, for our clients, the faster we can get stuff live on their website, the sooner we'll see SEA results, et cetera.
And so that's really worked well for clients where. They've got multiple products, multiple languages, multiple combinations, and
[00:22:12] Mike: so lots of SKUs, lots of stock keeping units. If I've got a thousand different SKUs and I want content for each of those in multiple languages, that's market context specific.
I'd imagine a lot of that work that was done by humans is now done by ai.
[00:22:28] John: Yeah. And then is, and then, but you still got to have. Real quick, the human needs to bookend it
[00:22:35] Mike: in the human and the lead. Absolutely. Yeah. Gotta brief it and review it.
[00:22:38] John: Yeah.
[00:22:38] Mike: Bookend needs a good example actually, John. Yeah, I get that.
Yeah. Yeah.
[00:22:41] John: And so then it might have been that we're still employing that same person, but their job has now shifted to being, uh, expert at prompting the tool, getting it set up, and then an editor at the end of it, et cetera. That's on the content side of things. The second side of things is also for us is code and tooling.
So. We've traditionally always built like tools in-house and we, we just spoke about one of the tool called Rose Writes that we use the content creation. Majority of that has been human coded, but we've now moved into this section where we'll get a client asking you a question or a particular problem and we'll go, well actually we're gonna build a tool that's gonna help us answer that.
And that tool can then be reused multiple times. So rather than a lot of. Analysis being done in Excel or manually. We're now going to tools like Cursor and Bolt and building very specific niche tools, which before you would've had to brief in a developer. The developer doesn't really understand paid search or SEO, LinkedIn marketing, et cetera, but you do as the expert and you are just basically conversing with this agent, which is a developer, and then hey, press and natural language.
In inter language and then you can go, no, that's not quite right. I want it like this. Or literally, sometimes it has been drawing something on a piece of paper, taking a photo of that and then giving it to the, the tool. And the tool does it. So we're moving a lot, lot quicker on building not mission critical tools because still, if you want anything that's mission critical to, you know.
Your business in terms of like checkouts or systems. You know, you have to get humans to code that in partnership with ai, but for these kind of smaller, more functional tools that will help you do your data to job quicker, that's where we're using AI to build them. And that also helps our and 'cause we're all over the world now.
We've got offices in different locations and that means. We've got this tool, we can send that tool to everybody in the world and then they'll follow that same process, whether that's like auditing, et
[00:24:44] Mike: cetera. Exactly. So it's repeatable, it's consistent quality. It's faster. So you are using the enabling technologies.
To re-engineer the way that you do work, and probably also the kinds of work that you can do, because five years ago you couldn't have done some of this stuff for clients. It wouldn't be possible, or it'd be extremely expensive.
[00:25:03] John: Yeah, yeah, definitely. And the other thing is we're also then taking people who had a little bit of technical experience, but were more like they were at S-E-O-P-P-C manager and now we're expanding their skillset.
And to be able to kind of move into that. So we've still got the same amount of developers, but those developers now are doing more complicated or they're refining the things that the SEO car managers and the execs can do, et cetera. And also in my team, we're now getting people kind of coming to us with, oh, I'm trying to build this thing.
They'll have a go at it and they'll then start to work out, actually, this is a good tool to work with or it's not. So the quality of the ideas that come through to us. Are a lot more briefed out. They're sometimes demos, et cetera. So that's really helping us. The big question in the future will be, do these tools help the account managers and the executives to be able to be so much more efficient that rather than them having five accounts.
Per team per head, they can then service 10. And so it probably won't be that we'll ever go. Right. We can make people redundant 'cause we don't need them anymore. It's just gonna be able to improve that ratio of the number of clients that they can service, which will make us more money kind of in, in the long term, et cetera.
[00:26:21] Mike: I think also John, again, in this book by C Childry, he was talking about a basketball example, old model you basketball players. Apparently you were like, you'd stand in a certain zone on the court and that was your skill. You'd practice being in that zone and you'd shoot from that zone and you'd get a high score rate.
And then with advanced analytics, they worked out that actually you need to be multi-skilled. You need to roam around the court. You need to, the value is in the coordination between players. Not being the sole expert, being the hero as it were. You've just described the same thing, is that people that were very deep in particular SEO aspects, now actually need to cover more bases of knowledge and need to be a little bit broader in the way that they work.
So we're going broader rather than deeper is what it sounds like as human skills, as human workers.
[00:27:12] John: Yeah, definitely. And I think also the idea is as well is that they could probably start to do things like, we've also started to do like YouTube briefs and video briefs. We're not doing the filming yet, but we're kind of, we've almost started to say right, we've now got a bit more time on our hands in certain areas, but also 'cause we've got these tools.
We probably didn't go near that before making scripts and making briefs. Now we can do that. Yes. That's happened a lot more is that we've actually breached out into other services, which would've been a little bit more, uh, that's gonna take a lot of effort, a lot of kind of learning curve where it's actually, it's a lot more attainable for us to be able to do these types of things now.
[00:27:49] Mike: Exactly. So I've been speaking to a lot of agency people over the last three or four months. And there appears to be a bit of a theme emerging, which is moving from performance marketing agency as a label towards growth partner. Does that ring true?
[00:28:03] John: Yeah, I think growth, definitely Growth or discoverability, we kind of call it.
And that's a little bit because of consumers will find out about you in lots of different places. There's, you know, there's, there's, you know, they may be on a social channel, they may be on search, they on ai. And then from our point of view is like where does the client spend their best? Budget to be able to be discoverable because they need to be on all these different platforms.
And then as we were just kind of saying before, that means that we need to have a wider skillset to be able to co cover all of those different basises. But yeah, I think what we see now is. That traditional SEO or pay surf model of 5,000 people per month are gonna be searching on this particular product.
It's actually, well, some people will find it here, some people will find it there, and we've gotta be visible across all those different areas. Um, and it makes our jobs a bit more complicated. 'cause now video content, um, we have to think about what's happening on Reddit, what's happening on these social channels.
And so yeah, that discoverability can happen across. Lots of different channels, as you were mentioning.
[00:29:05] Mike: So John, if I took a leap of faith or picked a scenario rather into the future in the next five years, so see if my thinking holds up. I think where we're heading is growth Partners will be multi-disciplined because the asset allocation model, if I'm a finance person, marketing have a budget, they've got a million dollars.
They need to optimize the way that million's spent in order to drive growth for their business. So they need people to work with who understand all of this. And can orchestrate it, can pull it all together to drive effective growth as opposed to, I need an SEO expert. I need a GEO expert. I need a paid media expert.
I need a social expert that's gonna start to collapse. Is a theory. Yeah. What do you think?
[00:29:51] John: I think, uh, yeah, I think there's some areas where it is going like that. I think the other thing is probably brands want to have coverage in all of them, so, and then they want flexibility as well. And so, uh, you know, it might be that we'll have a client who'll go right.
For the next six months, we really wanna push this particular channel, or we wanna launch in this channel, and so therefore, us as an agency need to be flexible. The other thing is for clients as well. Do they want to be talking to four different agencies, going to four different Q Bs, four different weekly meetings, et cetera?
Yeah, it might help them summarize that, but if they can find a partner that's able to understand the business problem and then work out how you execute it, um. They'll be much more inclined to work with that one agency. The other thing is also we're seeing as well, is that you've got your traditional creative agency who may create that brand assets, the videos, the master brand assets, the master kind of brand assets, and then it's gonna be going to the agency as well.
Yes, you're doing our PPC or you're running our affiliates or our display, you can use those AI tools to make a version of that asset as long as it's conformed within the brand stuff. And that's much more efficient. And also from a performance marketing point of view, we'll get better results. 'cause sometimes we'll get an asset that might have been.
For TV or good for a billboard, but it doesn't work in our, our different scenarios. And so that's where I think there's some efficiencies in allowing your digital marketing agency, performance marketing agency to, to increase their remit because they have these tools, these AI tools available to them.
[00:31:23] Mike: John, what a brilliant episode. Really, really, really good. Thank you. Gimme a couple of takeaways. And you can take it from whatever perspective you like as an agency, as a client, as a kind of search expert, what are your kind of couple of takeaways from this recording?
[00:31:38] John: Yeah, I think the first one is that you've gotta start to really investigate how visible your brand is on AI search.
Um, you know, where are you appearing, where you're not appearing? You'll see a lot of similarities from your SEO work. So if you've been good in certain areas in your SEO work, you should be good in, in a lot of the geo findings. But what you'll, you will find is. There will be these pot of websites that are having influence on the results.
So these will be the comparison websites. They may be listicles or, or directories, et cetera. And you may have been thinking they didn't have as much importance in the grand scheme of traditional SEO. They're now having influence. So definitely look into those. And there's a lot of tools out there that do this, but one I would recommend we don't own it is, uh, peak ai, which is a nice kind of starter level tool, which does scale.
Um, but that can be a nice little tool to get set up, run 30 prompts, but also there's prompts. And not your traditional SEA things. 'cause when people are searching on AI, a lot more lengthier in their prompts, they'll probably put their own particular use cases in there. You know, I'm a small to size medium business, I'm looking for some software to do X, Y, and Z.
So when you build out those tools, have that into consideration. And then, yeah, on the second side of things, in terms of using ai. In your marketing and, you know, processes or like if your market is internal or with an agency. I think for there is, there is a place to use it, but you need to decide where that place is, so where you know you need the human interaction.
So if it's in the content. Writing side of things, yeah, that's fine. In the middle, but not in the editing, not in the brief making. Maybe do you want AI to be making your weekly reports? It might not have the full context of everything that's gone on, but you know, if you make a graph and it's got, you know, you need a little commentary to go for it, maybe AI is good enough to do that.
So. Yeah, on that second point is just you've gotta test and work out where the AI is useful and where it's actually gonna deliver less value. And if you're service selling your time as a service, you need to make sure that it's not gonna devalue the service that you give, uh, in the long term.
[00:33:47] Mike: Absolutely, and that'd be my kind of add-on at the end would be look at your commercial model really carefully first, and then look at how you're going to adopt AI and make sure you're not giving away all that value to your clients.
John, it's been amazing. Thank you. Where can people find out more about you?
[00:34:02] John: Yeah, so go to the website. We are raised.com, or just have a search on LinkedIn for John Campbell Roast, or John Campbell SEO. I'm sure you'll find my details on there.
[00:34:12] Mike: Brilliant. John, it's been a real pleasure. Thanks ever so much for joining me.
[00:34:15] John: Thank you.
[00:34:16] Mike: Thanks for listening to Higgle, the B2B Sales Club podcast series with your host Mike Lander. Please subscribe so that you'll catch all the next episodes.