How To Create a Revenue-Driving AI Search Strategy for 2026

Heat. Heat. [Music] Hello and welcome to the dojo research marketing podcast by Exposure Ninja. This is episode six of our digital marketing strategies for 2026 series. And this week we're going to be talking about how to create an AI search strategy. I'm Dale Davies. I'm the head marketing expert. I'm joined by Charlie, our CEO. Charlie, why for those who are tuning maybe for the first time really aware of what AI search is and for maybe a reminder for those busy marketing leaders who are maybe heard of it, just need a bit of a reminder. Why is AI search important? So AI search is searches within LLM, large language models. Chat GBT, Gemini, Perplexity, Claude, Google's AI mode. The list goes on, but those are probably the most popular ones at the moment. And there's been a bit of debate in the industry about whether optimizing for all of these LLM is different from SEO. And my opinion on that is it absolutely is. these answer engines, generative AI, LLMs, chat bots, whatever you prefer to call that ilk of chat GBT and similar competitors is a different environment to optimizing for Google. And more and more people over time are adopting ChachiBT and other LLMs to make searches. And those are both searches in the sense of they're doing research for products and services they're in the market to buy as well as using it for other tasks for those very early research phases top of funnel stuff just asking for definitions of things and the generative uses as in things like writing emails or doing tasks or those bits and pieces as well. So there's quite a few different kinds of use cases of AI. And if we think back, if we cast our minds back all the way to say December 2023, January 2024 time, chatbt and other LLMs were just not on our radar. Like they hadn't come on to the scene. And then it was around April 2024, that sort of spring summertime, they really started to explode. By the end of 2024, suddenly everyone and their cousin had Chachi BT's app on their phone. And that trajectory has just continued to increase over time. Now, one of the big debates is okay, CHBT and other LLMs. It's continued to increase over time. Huge amounts of people using them. Around 2.5 billion searches per day on ChachiBT. Compare that to Google. Google still has around 8.5 billion searches per day. It's still the market leader. They're still the most dominant. They're still the most wellunded even though there's plenty of funding going into OpenAI, Chachi PT's parent company, and all of the other LLMs as well. So, that's the big question at the moment for many marketing leaders is knowing what we know about the rapid rise of ChatBT and other LLMs. Now, how much of our budget and time and marketing strategy are we investing into AI search optimizations? And how are we balancing that with the traditional SEO strategies, the traditional search strategies focusing on both organic Google and paid Google ads as well? Just coming back to your point on the balance between dedicating time to AI search and SEO. Are customer is customer behavior changing at the same kind of rate? Are we seeing like a lot of people switching over to using solely Jack GBT versus Google or is it people using a combination? Like has the buyer journey changed? It's hard to know exactly, but what we do know is that people tend to use both still. It's not the case that people have necessarily completely given up on Google and now they're only using chatbt. A lot of the time people use chatbt to do a large part of their research and then might also go to Google from there. But the trend that we are seeing is chatbt integrating links and buyer journeys more directly within its own answers. So when Chat GBT first rolled out, we were really just getting textbased answers, sort of what you might call a blog post type of reply that was a bit more conversational, a bit more engaging with prompt questions of what to do next. As time has gone on more and more, we're actually seeing chat GBT link out and we're also seeing increase in partnerships for Chat GBT. And when I'm talking about partnerships, I mean all sorts of different things going on. So, ChachiBT has had a partnership with uh Open Table, for example, for making restaurant reservations, which you can do in its agent mode. Um, it's now got a partnership with Etsy that it announced just a couple of weeks ago for an instant checkout where you can buy directly through ChatBT for products on Etsy. more and more we're seeing people be able to make purchases directly out of chatbt whether they're referred out or whether they're actually within it and this is kind of at the very early stage of purchases being within chatbt that's what we're seeing with Etsy there's a high likelihood that Shopify is also going to follow suit with that and be next so the buyer journey is very much getting compressed for people who use LLMs like chatbt because now they can just do so much of the journey so much more quickly within AI rather than having to troll through the 10 blue links, all of the Google search results and do all the research themselves. ChatgBT just gives them the answer much quicker. Is it the same for service based businesses as well? You mentioned there like a couple of partners that I would assume were more for e-commerce or retail where I'm doing a bit of research for like a a product line. I'm looking for trying to think of something that might be on my desk. I'm looking for a new a new case made out of uh sustainable materials made out of uh you know reuse plastic things like that. But sure I we're going to do that kind of search on chatbt or maybe one of these other platforms. Do you think that if I was searching for I always come back to it mortgage advisor for example I would do like those kind of searches on people are doing those kind of searches on the AI platforms too. Yeah absolutely and financial advice is a huge area that people are using chatbt for. as many podcasters and YouTube YouTubers talking about how actually rather than getting a financial adviser, they've used CHBD to get that advice and then as part of that journey decided, oh goodness, I do actually need someone qualified to give me advice or I do need an accountant or this is the kind of mortgage broker that I'm going to go with or you know, I'm looking into these savings accounts because chatbt and other LLMs, most of them can search the web. So if they are doing something that requires very recent information like looking for the best mortgage rates, looking for the best savings accounts rates, whatever that may be, that can be done within CHBT and it's actually very good at doing comparison pieces. So what we're seeing in those kinds of service-based journeys or lead generation journeys if you're thinking from the business's perspective quite a lot of the time a lot of that research stage is now happening in chat because it's much more conversational and it's much more personalized whereas in the uh in the olden days or if we say the traditional way of searching on Google it's actually left for the user to have to read multiple different websites in order to do the comparison. themselves and that might be looking for, you know, something versus something like a comparison type of article online. But a lot of the time it was actually reading each individual website looking at mortgage brokers. Then we had the advent of course of things like uh compare the market and go compare which made life a bit easier for everyone. And now we've got the advent of AI that can actually personalize the answers based on the exact information it's being given by the user. So I think we're just seeing more and more the buyer journey trying to be made easier, quicker, simpler for people who are comparing services that they might go on to purchase. I have to say last night I was doing a bit of deep conversation with Gemini and be like you know here are the things I'm going through at the moment. and I'm looking for a bit of advice and it found me uh it recommended some local options to me like three different uh specialists um to reach out to and said like these people would be great cuz it understands like what these people are good at and what my personal set of circumstances are. So rather than me having to just troll through loads of pages as you said like opening up a directory like check a trader or something like that. It's just a little of the hard work for me and I've raced through to that lead capture point of filling out a form and saying please call me much much faster and it's just saved me a whole bunch of time which is what the biggest selling point is of a platform. Oh, and just last week I used it to compare different pension providers to open a SIP and then made my decision off the back of advice from ChachiBT as well as advice from external advisers, other people that I'd spoken to as well. But it was very very much part of my buyer journey into deciding which pension provider I wanted to move to for the long term. So it's 100% used in all sorts of decision- making, financial decision- making, but service-based decision- making that people need both at work and in their personal lives more broadly is happening within chat GPT as well. So it seems like it would whether I'm a market leader or a challenger brand, I need to be really sure that AI is referencing my brand in the best way possible. Yeah, 100%. And I think that takes us almost down to the what is the strategy for AI search? What what can you do? And the first principle of that is understanding and accepting that you can make changes to influence AI search. There's a lot of conversation going on about whether AI will just eventually pick you up if you keep doing good marketing. Well, maybe in a lovely ideal world where everything is rainbows, rainbows and unicorns, it might be that AI just picks you up and you've done a really good job. Particularly if you've done really strong PR campaigns, you've positioned very strongly in the market. But actually, this is just another way of saying if you do the right things and optimize in the right way, and if you happen to have been doing that already, lucky you, you're ahead when it comes to AI search. If you haven't, you actually do need to employ tactics to improve your AI search visibility. And what you're touching on there, Dale, the sentiment as well, like how positively or negatively your brand, your business is spoken about within AI, how it gets compared to competitors and other options in the market as well. And that comes back to actually first the belief that you can optimize for chat GBT which our own research our own case studies our client work has shown you you absolutely can drive leads and revenue through optimizing for LLMs and that that isn't the same as just doing SEO as we know it in the sense of optimizing for Google. That's actually one of the biggest things that kind of uh bothers me when there's the comparison between SEO and AI search optimization is people looking for trying to say that they're they're one and the same that you're optimizing for Google, you're also optimizing AI platforms. But it doesn't seem true at all because like even the metrics of how you measure performance with AI search are completely different and don't aren't impacted by some of the traditional stuff you might do but there is that underlying foundation that you need as you mentioned there about SEO the technical side as well. Is that correct? Yeah absolutely. When I think about how we optimize for AI search optimization, the first thing to think about is it's different from Google because it's not actually Google. It's not Google's algorithm. So depending on which model you're optimizing for, if that's chatbt, Gemini, Claude, Perplexity, etc. They have different underlying search engines that they're using. So there is part SEO that is based on the kinds of results those search engines give. But they also have their own underlying training data. And unlike traditional Google algorithms that are ranking a sort of top 10, if you like your page one, AI doesn't need to do it in that way. It can scour search engine results across multiple multiple pages. And we know that sometimes it references from page 2, three, four onwards if it feels like that content's more relevant to the user. So first thing to think about is that those they're different. which ones are you actually focused on optimizing for? For the majority of businesses, that tends to be trackt because it's the most dominant. There's exceptions particularly in B2B. If you know that your target customers use uh Microsoft, they have to have co-pilot. That's their one that they have to use at work and that's mostly the type of business you sell into. You might be focused on that. It varies. When it actually comes to the AI search optimization, for me, there's three very distinct phases. And the first one is the technical side. Similar to what we think of as technical SEO, all of those things about having a good functioning website that is healthy also apply to AI. Like Google's not going to show your website if it has loads of errors on it. Neither is AI. The patience of AI is probably even less arguably. So things like having schema in place is incredibly important. making sure you don't have 403 errors, 404 errors, all of your redirects clearly sorted. All of that technical side needs to be in good shape before you can start going even more deeply into phases two and three, which are looking at your own website content and how you're optimizing it and then looking at your digital PR and citation building or content that you're doing off site that gets referenced elsewhere. So you sorry you mentioned there about DPR and citation building. Is that one and the same as like earning back links and things like that or like in fact are the back links as valuable as they were with SEO or are we looking to go about things a bit differently? We are definitely looking to go about things differently. So, in SEO, we think of link building or backlink building because we're trying to get specific text and website URLs hyperl on other websites because that's what's most important. When it comes to AI, the link doesn't matter. What actually matters is the context at which you're being referenced in, all of the text around it. So all of the context it's being given about the business and the citations or the sources, the websites that AI likes to reference are very different. It's not using the same algorithm as Google to decide what's authorative or not. They have different preferences. We see different types of content linked. And then the other bit that's also really different is that AI gathers information about who the business is is across so much of the web across articles across your descriptions in your social media channels, Reddit, review forums, things like Trust Pilot, G2, depending on the business. Uh I've even seen specific ones in the wedding industry like for hitched. I've seen specific ones um outside of the wedding industry, but in something like housing and solicitors with review solicitors, like totally different websites depending on your niche, depending on what it thinks is important, get scraped, get referenced by AI. And it's also taking that kind of context around how you describe your business and the positioning statements you make about it or we could say the USPS, what you're known for, what you do. It's pulling that across different areas of the web. And that can mean that if you haven't taken control of that strategy and deciding how you want to be seen, you may be being referenced in ways that you don't feel are accurate for your brand. And that's because there's a huge amount of weight given to digital PR and other publications in AI. Much much more so than the weight given to content on your own website. Does that mean I should be dep prioritizing my own website when it comes to content creation? Should I be producing less and just trying to get content produced elsewhere and published elsewhere instead? I think you need to do both. You can't just again it comes back to there being three phases, right? Technical, your own website content, and then digital PR and citations. If you just try and like elbow out one of them and knock it off and focus all your eggs in one basket, you're still going to struggle because you've not taken care of all three of those phases. And it's almost simpler, I think, to start with once you've got the technical done to start with thinking about your website content and making sure you're described how you want to be known that you literally say what you do on your website, which sounds basic, but actually a lot of websites have got so into the like trying to write cool and trendy language everywhere that they sometimes miss just saying exactly what it is that they do for their customers and making it clear. If it's not clear on the website, it's going to be very difficult for AI to understand. But the website is also particularly important for any branded type of search. So any search that goes on in AI, specifically using your website's name. So the the bigger the business is, the more recognizable your brand is, the more important that becomes because you're more likely to have higher volumes of branded search and you want to make sure that you're known in the way that you want to be known. So, can you knock off website content? Like, I definitely wouldn't because it's the first it's the first source of truth, right? It's the first place you have to optimize for. Then you've got a lot more then you've got a lot more you can do offsite once you've got that in place. I was looking at the rankings of health insurance companies in the US last night and there's one company or maybe a couple of companies who are just dominating AI search versus um traditional SEO where they're doing well but not nearly as as well as you'd probably be hoping for. And I think part of that is to do with you how much on-site content they have which the AI platforms are citing but it's like that external stuff as you said before like how people are talking about that particular business like I can't remember the name is Kaiser something let me check my notes Kaiser Permanente I guess that's who it is um they're like doing really really well they've got like 74% coverage when it comes to AI uh visibility and how many the times they come up in the prompts and and the results that come back and again like they have a lot of on-site content that's helping but it's how people are talking about them in other places like Quora and Reddit and you know elsewhere I do comparisons between them and say United Healthcare and and so on. It's like you know how the customers are referencing them as much as how many pages they've written about this type of insurance plan versus another. Well, and this is why both are incredibly important. There's also a lot of content that you would put on your own website that you wouldn't put out in PR. So, for example, I was doing some research recently for uh wedding venues. I'm not getting married in East Sussex. And as part of that, you see people are going to write in prompts that are like, I'm looking for these kinds of wedding venues. Barnes, whatever it is, country houses. You're giggling away. Stop it. Barnes, country houses, fairy tale castles, whatever it is they want for their wedding venue in East Sussex. But the other thing they're also going to write is their budget because very few people are going into their wedding with unlimited budget. So, say you're then putting in a budget qualifier like max 20K on my wedding venue, then the kinds of content that gets referenced, the the top cited or the top ranking, I don't like to call them ranking, but most visible within the AI search results were actually pulling budget information from those websites. So, the actual venues who had said, you know, this is our package. is whatever it is, £8,000 on a weekday goes up to £14,000 on a Saturday, Sunday, or a bank holiday. But you're never going to write that in your PR articles because over time your pricing is going to change. People are not going to be writing that they're this is the total cost of a package for a wedding in PR that you're doing offsite. But all of the wedding venues that had written those types of packages, those types of costs or something similar on their website were doing much much better in AI because there was that qualifier in a lot of searches around budget and pricing. So AI is looking for that information on their website. you mentioned in there about you wouldn't describe it as ranking may instead as AI visibility or brand visibility but there are tools that I know that we use with we're partners with a few of these you know AI search uh tracking uh tools where they do reference like position as in the average position when a link is citated um so maybe the 1,500 times they've done that prompt or prompt around that topic it tends to be cited did first or third or something like that. Is that a key metric? Um actually a better question maybe what are the key metrics that we should be tracking prior to creating our strategy to see if we're matching our objectives. So the reason that I hesitate on using something like ranking or positions even though there are clearly positions in searches and you you clearly want to be in the top positions is that it's not keyword related in the same way that Google searches and it can get very messy if you're trying to specifically tie rankings or position directly to specific keywords because prompts are so much broader than that. And at the moment there are tracking tools out there like you mentioned Dale that have positions or estimates of positions which is fantastic because the more data we have as marketers that's amazing for us to have something some semblance of being able to track it. But at the moment the prompts that we're tracking we don't know the actual volume of searches of those prompts. It's kind of synthetic data. It's data that's estimated and it's using a lot of Google's keyword data underneath that to give those estimates as well. So that's why I'm saying it's not quite the same as when we think about Google rankings even though there is a clear hierarchy going on of positions within AI searches. um metrics more useful to measure. For me, it always has to be the actual traffic and conversions that you're getting through from LLMs. And you can measure that within your Google Analytics and that will show you the referral traffic coming through. It will show you anything from those platforms that converted. You just need to separate it out. If you're e-commerce, then you've got revenue set up in G4 as well. You can also see the revenue driven by AI. Um, we know of course that lots of people will perhaps do a search in AI and then copy and paste and search something in Google. So, it won't be aund 100% accurate of all of the touch points in AI, but it'll give you a really good starting place and it's probably the most concrete of the metrics you can measure. Then there's visibility scores, how visible you are over a certain set of searches or prompts, which is probably more useful to be using rather than thinking about trying to pinpoint it down to specific rankings at the moment. Citation rate as well, how often you get cited in answers, how often your content gets cited. share of voice, who's which is pretty much who's who's winning in the market. And that is very important for businesses who are in particularly aggressive competitive markets where competitors are jostling very very closely. Then you want to be knowing who is winning with the share of voice for certain topical areas. And then the the really new one is probably sentiment score, like an idea of how positive or negatively you're spoken about in AI. I suppose that's impacted through your digital PR and and your onsite content too. Like how well the AI platform is going to talk about you is going to be based off maybe not your own voice and the things you say about yourself, maybe what other people are saying elsewhere, like on Trust Pilot for example. Yeah. Yeah. And I think we see particularly when it comes to sentiment significant amounts of references to review sites and also to forums like Reddit. And from time to time you also see social media type of content coming up as well. And so that sentiment is really much more down to third party sources and what customers have been saying about your business online. one of the huge reasons that actually managing your reviews effectively online is so important for AI search as well. I feel like we're getting closer to the topic of the types of content that need to be created. So, we've got these different options and where we place it. So, we can do our own website. We can work with partners. We can work with all the publishers and we can do our best to kind of you know assist our customer service teams as well to deliver a great service which then should turn into great reviews. But what kinds of content are best for marketing leaders to focus on with their teams and to strategize around that are going to be influencing the A platforms but also going to be cited to. Yeah. So there's really two parts to this question. The first segment of this question is what types of content in terms of the format and still by far it's text content that gets referenced the most. AI generates text and so therefore it uses that. That doesn't mean that AI won't reference or use other types of content. Sometimes it will. Sometimes videos for example get pulled into AI results but predominantly text on web pages and blog posts is the main type of content that gets referenced both from when we're thinking about our own website and when we're thinking about third party websites as well. So firstly that second segment of this is the type of content in terms of what are we actually writing about and where do we want it to be. So there's lots of different types of content that gets referenced by AI. You can you'll often see even just service or product pages being referenced by AI. That example of wedding venues I gave just then that was literally a pricing page on someone's website. So for that you need to just understand what kinds of searches your customers are likely to be making and therefore what kind of answers they want to see. If you can answer some of those questions in your own content, great. When it comes to the PR side, the best way to do this again is having an idea of the kinds of prompts you want to show up for for your business and then actually looking at the articles that get cited. So you can do that manually by literally typing into Chat GBT whatever prompt or search it is that you want. Then you can open up the panel that shows all of the sources. It will literally give you very similar to like an academic essay with footnotes references to everything that it's cited. So from there, you can do it the manual way. You can do a couple of different searches, the things that you care about. You'll probably see some similarities in the sources or the websites, the articles that get cited. Then you can pull all of the topics, do an analysis of the kinds of topics you want to publish on. And you can also use those specific sources to contact the editors, contact the journalists, try and get included in their articles. One of my favorite examples has been um comparing CRM systems. So if you're comparing like HubSpot, Monday.com, Salesforce, all of those types of CRM, you'll really really frequently see like articles on which is the best CRM in 2025. So loads of trends articles that are very recent looking at what kind of integrations they will have and actually doing comparison pieces like HubSpot versus Salesforce versus Zoho versus Monday.com and so on, which gives you a really great inclination, right? If those are the 2025 trend pieces now, and we're already in October, you can guess that the best PR articles to be in next year are going to be 2026 comparisons that are very similar. And that's exactly the kind of content that journalists and editors are going to want to put together. If you're thinking, Charlie, I like dream of the days where I would have time to do manual searches in chatbt and make a spreadsheet. I do not have time for that. I totally understand and there are many many tools that will do this for you at a much larger scale. So those are tools like PKI, Scrunch AI, Profound that can actually you you preload prompts. So say you preload 30 prompts that you know that your customers are quite likely to use. They will scour across all of those prompts, pull all of the sources out so that you've got a list of exact URLs, tells you which ones you're not mentioned in and which ones your competitors are mentioned in so that then you can prioritize the ones you want to go after. And then you know what? You can actually just copy and paste those and ask AI to do an analysis of the trends for you. Just plug that into chatbt and say, "Hey, what kinds of topics are mostly coming up here? What would you recommend I do from there for building links from here? Yeah, I've got one example open right now from my research for uh health insurance from last night. And you know, some of the prompts you can put in there are kind of fairly expected. You know, which is the best health insurance currently, but you can really go to town on it and really you get really descriptive on on on a particular um buyer within your, you know, target market, within your target audience. So, you know, one of the examples I've got here, I actually pulled from Kora or Reddit. I can't remember. I went searching for like actual questions that people are searching for and you can get really specific and say you this one is my work my wife works at a hospital and has 119 which covers the family. I work at the post office and I have my own self plan. My wife's job says if my job offers healthcare I must take it. Is this true? Like you can get really specific about like how do I share health insurance across a family rather than oh which is the best health care for families or families of two families of three. Another example here is uh which is the best medical insurance plan for senior citizens who are parents aged 75 and 80 years which also gives it deduction. This is a very hypersp specific search which somebody actually has done. Well, first of all, this was proposed on like Reddit. Somebody was asking a community, asking a bunch of people who've been through it. People are now sending those same searches to AI instead. And these tools are just great for kind of helping you to track. Okay, maybe not too too broad, but like some of the core, you know, header terms as we used to say in key in keyword research, the broad, you know, the top kind of stuff, but you can go slightly wider towards that like super long tail uh as well and track that to see roughly how the parents of let's say one of our uh our clients does bathroom installations for the elderly of people require them. uh it's often not the actual people who need them who are doing the searches, it's the children of those people. So their searches are very different. It's which is the best one for my parent who has dementia and and so on. It's so completely different and it's good to kind of track again not too too broadly what the results are of those searches. Yeah, absolutely. And I think that's a great example of understanding the types of customer that you have before you embark on an AI search strategy because otherwise you are going to find that you go too broad and it feels very overwhelming. It's better to actually focus in on who are the customers that I want and what do they specifically search for my business. Segment those and then actually put in prompts so that you can track them long term. I realize we've been talking an awful lot about the organic side of AI search. Is is there a paid site of AI search at the moment? This is like the big question on everyone's lips. Whenever I present about okay, how can you manage AI search organically? People are like my goodness this is this feels like a lot of work and it kind of is. And so the next question becomes like are there going to be ads? Which some people like the idea of and some people are like wow AI is going to become really scary if there's ads. So, at the moment, the only ads that are really running are in Perplexity, and this is in the US. It's a pilot group that's been running since 2024. And when I say a pilot group, it means they're closed off. You can't the average business cannot go in and start running ads. And this is because tools like Perplexity, whenever ads get tested or put in beta version, they always want to do it with big brands. So in that pilot group, it's the likes of Whole Foods, Universal, Macan, PMG, indeed the recruitment site as well, very very big brands because they're more likely to get great results that then they can report on and say, "Look, our ad platforms get these percentages. It's fantastic results." So yes, there is definitely ads coming into AI. The dates at which that will roll out, we still don't know. Probably the biggest question on everyone's lips is when is it coming to chat GBT. So open AAI who are ChatgBT's parent company Sam Alman has been saying okay I don't really like ads I we don't want to have ads. He was verminently against ads. This was a few years ago and over time that rhetoric that language has become softer and softer and now open AR at the stage where they've publicly said that they are exploring ads. That stance is so soft that they're actually exploring it. Plenty of reporters uh estimating then that that means there would be an ad roll out in 2026 that they're recruiting for the kinds of staff now that they would need to actually integrate ads into the system. And this is probably not surprising because AI at some point needs to monetize at scale. So running ads is going to be the next kind of logical step in that path to them generating their own revenue. And I guess they're eventually dreaming of profits as well. Yeah. I just think back to Google where it's been, you know, a brief period of time of just being this great search engine just you completely different to everything else that was available at the time, Yahoo and AOL and everything. And it wasn't until a year or two, three years later where they added in Adwords and you could sponsor those kind of results above the the classic template links which feel like ancient history at this point. Um, it seems inevitable that next year, maybe the year after, these ads will come. I mean, they have enough money to just keep running without doing any like strong advertising just yet. I suppose you referenced it earlier, this instant checkout that they've added in where they're taking a percentage cut of the sale as well. I don't know what the the number is. It's 5% 10% but they're taking like a little portion of that will be one of those like first revenue points. Yeah. And I think ad experiences are probably going to look quite different within chatbt and LLMs as well. They're still trying to figure out what it looks like. They don't know exactly. But we also know for example that um Chachi BT are integrating apps more and more. They want actually people businesses putting apps inside ChachiBT as well. In its current interface, it's it's quite difficult to see how that's going to look long term. Like at the moment, it's almost like a very commandbased interface. If you remember video games from like the early 2000s, that was the kind of thing that was going on. It was like you were putting in commands in. And actually, it's a really impressive growth for quite a simple like the AI is not simple. It's fantastic, but it's quite a simple user interface that's going on. It's literally just like a search bar and a chat screen. So, longterm, I expect actually we're going to see quite big differences in the whole design of everything that is inside an LLM. But at the moment, the question is, are they just going to be like in text? They're going to have to be marked as sponsored because there's law. There's regulations that go on about how ads work. and what's that going to look like? Is it really going to look like it's that sponsored? How will users definitely know? There's so many questions at the moment. So, I think there's probably going to be quite uh a lot of change in the design, in the setup, in how results in LLMs look at the time that we actually start seeing ads as a general general user population. You mentioned that there's going to be a continued like change in the look and feel of ch GBT in particular and AI platforms. Is the functionality going to continue to change too, do you think? I definitely think so. Already we've seen models becoming more and more advanced at a pace that's even quicker than I had originally estimated it would do. Seeing functionality like agent mode for example in chat GPT where an AI agent does tasks for you hands-free version it does everything in its own virtual browser that is already extremely different to those early versions of chat GPT that we were seeing. So whilst it may look like not much has changed in kind of the design or the UX interface, actually a lot has been changing in terms of its capability and remember many of those early early models they were just using training data. They weren't even searching live web results. Now so many of the searches are using live search results, live web searches in the background of them as well. There's so many different modes and models coming out that I definitely think we're just going to see even more technical advancement over the next couple of years. I know you've referenced it on previous podcasts uh but for those who may be listening for the first time, perhaps you can give an an example of how that aentic um tech has been put to use in some of the searches that you've been doing. Yeah. So for me, I've been using it for a lot of shopping. uh the kinds of shopping that I don't want to do, for example. So, I've had it check out for me to buy uh coffee subscriptions. I've had it buy uh cat food with like an urgent kind of date. So, you can put in something like, I need a kilogram of cat food and I need it to arrive in the next two days and I want it to be premium quality. Then, it will go away and do all of the searches for you. It will check all of the delivery times. It will find premium quality cat food. Then it will bring to you in a virtual browser the checkout page ready for you to pop your card details into and go without you having to search that in Google, go through all of the different results, check all the delivery times to then figure out which one you're going to buy. It does all of that work for you. But there's loads of examples and it's not only checking out on e-commerce of what agent mode can do. I was speaking to someone who uh runs a consultancy business the other day and they told me that what they were doing was when they had a a client project come in and they needed to create multiple different reports. They would actually set the agent mode, lots of them, to go away and do different parts of deep research, bring them back fully written reports that they could then use based on actually going across the web, reading loads of different relevant context, and then they would be able to pull those all together to actually produce the piece of work that they were consulting on in the end. So there's loads of different uses for how people are actually implementing using agent mode in their day-to-day. Uh do recall that when I was doing some research for one of our videos earlier in the year around Rocket Mortgage, I used deep research within Chachupti and Perplexity to say, "Hey, go out there, go through all the earnings reports, go through every interview that their marketing team has ever done. just tell me everything about like what they do in terms of where their priorities are with ad budgets versus or particular channels and so on. But um it does bring us towards kind of the the main topic of this video which is how do we create a revenue driving AI search strategy going into 2026 when we know that this should be probably a priority for us. I'm starting from scratch. Where should I dedicate my time? How much time or money should I dedicate towards this? How do I go about doing that? Great. For me, the first place to start is understanding what you've already got coming from AI. And for most businesses, there probably is already at least some traffic coming from AI platforms. just dig into your Google Analytics or get one of your technical team to separate it out in Google Analytics so you can see what you've got so far. That will give you that early feeling of motivation of a win of seeing that actually something tangible is coming from AI. That's going to give you the confidence to explore AI search further. And one of the great things about AI search is this is still early days. Like the days are passing, but it's still early days. If you're actually going to start implementing an AI search strategy in your marketing now or in 2026, you are gonna have that first mover advantage where many competitors actually haven't quite got to the stage of implementing AI search even if they also have it in the back of their minds as something they need to do. So start there then start with okay what kinds of searches are people likely doing about my business within AI within chatbt within whichever target LLM you're deciding to do and I think that's then the biggest part the biggest piece of work becoming right once I have an idea of those prompts you do actually need to create a plan you do need to create a strategy for how are we attracting the right kinds of customers to then make a purchase. Whether that's going to be directly in touch, whether it's going to be coming through AI search, clicking, linking directly to your website, whether it's going to be copy and pasting things after they've done all of the research phase within AI. And that plan has to be broken down into those three phases. The technical is everything technically sound about your website that is going to benefit you in beyond AI search that's going to benefit all of your online marketing anyway. So definitely do that. Then it needs to be what kind of website content changes and a lot of the time it's going to be changes but there's also going to be updates. There's also going to be new content needed. What are the opportunities? What are the gaps? What are the items that need changing? what is the positioning that is unclear about my business on my website. So, the on-site content part and then the third part is going to be the digital PR and citation building part. And I'm including things like looking at your reviews, looking at Reddit threads, Quora threads, looking about if you've got a Wikipedia article about your business, thinking about your positioning, how you want to be known, which articles are being cited, and unpicking that so that you have a really, really clear list of target publications and target topics that you're going after in your PR. I just putting myself in the shoes of a marketing director, CMO, and thinking, "Wow, this that seems like a lot to do." Like I I I still kind of not sure on how much I should dedicate to this. Like SEO has been great for, let's say, I mean, their position SEO has been great for us for years. We added in a few more channels and things are working, but it's been like a slow burn. Should I slowly be working on this like just gradually or should I go like all in as much budget as I can throw into it? My opinion on that is it depends what you're seeing at the moment. If you're already seeing increases in AI traffic come to your website, that's a very early sign for you that there is something there worth optimizing for. For me, the biggest danger is the burying the head in the sand of, well, what I'm doing now works just fine. So, I'll just do more of that and ignore AI and hope it goes away. But this is not going away. These are different channels from Google. And optimizing for them can bring you traffic. It can bring leads. It can bring revenue for the business. And if you don't start doing it, you can bet that a competitor will start doing it. So it would be a very scary place to be I think to be ignoring it. The other thing that a lot of um businesses are seeing when they've been investing very heavily in SEO is the advent of AI in terms of Google's AI. So AI overviews has meant they're actually seeing drops in website traffic. They may be seeing decent impressions still, but they're not actually seeing as many clicks through. They're not seeing as much organic traffic as they once were on their website. Though the most important metric is conversions, if you are sitting on your hands and watching that and taking no action to put an AI search strategy and implement AI search tactics for your business, you are pretty much watching your business's demise happen without taking any action. And that's an even more dangerous place to be. So for me, when it's how urgent is this, it really depends on the business where their traffic and revenue sources mainly come from in their marketing. And I'm not suggesting that every business suddenly scraps their their Google SEO plan or their uh paid ads plan or whatever else it is that drives that investment uh that drives that revenue for their business. but actually spreading the investment so that you've got a percentage of budget that you're investing in AI search and that is highly likely going to be a very important channel for you if traditional SEO organic Google traffic has been a very important revenue driver conversion driver for your business I can't help but think of the questions I've had the last year and the year before and the year before that which is do you think that these AO platforms and chatbt in particular have an opportunity to continue at their current growth rate and earn you know a a larger share of you know search in general. Do you think that they can catch up with Google and impact their volume of, you know, audience or or users or do you think that Google may just start to claw things back with the introduction of AI mode and and additional tools like that? Well, it's the the milliondoll question, though. Is are AI like Chat GBT going to be able to outpace Google? What we know is that Google will not just sit back and allow ChatBT and other LLMs to outpace it. The release of Google's AI mode was an aggressive move back from Google. They are integrating, they have been integrating AI into their search results long before chatbt became as familiar to people as it is today. So Google are definitely not going down without a fight and they are by far the most dominant. They're well funded. However, if they don't make some big moves with AI mode and how they use AI, I think we could definitely see a world where LLMs become the more popular choice. And particularly, the reason I say this is particularly because, you know, old habits die hard. the millennials, the boomers, we're gonna struggle to come off of Google search. It's what we've always known. It's what we've come to trust. Even though it's got ads in it, we trust that. Gen Z, they do not care about using AI. That it is their normal. They grew up in this age. They are the biggest demographic users of chatbt and LLMs for searches. They're happy to do their searching on Tik Tok. I don't think there's a boomer that would dream of doing their research on Tik Tok for buying something. But what I'm saying is generational changes. The habits shift. We already have different search habits at one end of our demographics and the other and those shifts will continue to happen. So if we think, well, I I love Google search, nothing's going to change. That's a very naive way of looking at what's actually going on across the population as a whole. And there's been plenty of projections already. Seamrush did a great projection which was earlier this year about when they think LLMs including Google's AI will actually overtake traditional organic search and I think that estimate was around 2028. don't know that it couldn't be earlier based on how rapid rollouts have been. And do I think Google is suddenly going to die a death? 100% absolutely not going to happen. But I do think there's going to be more competition in how search works. It seems to me that regardless of whoever the top two or three are, the technology changes. Therefore, your approach needs to change and you've got to act accordingly. Well, and you know, we've we've had this uh situation where Google's pretty much been the most dominant. Like, okay, some people use Bing. Sure, some people use Bing, but now it's already been the case pre us thinking about chat and LLMs where we know that people searching on different platforms. They're actually search on Instagram, they search on YouTube, they search on Tik Tok, you know, some people search Reddit for example. This advent of LLMs is not that different in the sense there's more search platforms and Google had it has had glory days of being the main search engine platform. Now there's just more options for search in the market rather than just the the the sort of one be all and end all of where we search. If anything, I think it's good for Google to have a bit of competition like this. I'm going to bring us to an end with just one final question which is if you are speaking directly to a marketing leader right now what would you want their main takeaway from this to be and what should they be communicating internally to their teams whether that's stakeholder or the team that they oversee I would hope that a marketing leader listening to this has taken away that they need to start making moves on AI search they need to have a plan and then they need to be ready as a business to start implementing that plan and depending on the type of business the level of business and I mean level in terms of if you're in an enterprise business versus if you're a mid-market versus a smaller business the levels of buyin that you need to go through to make sure that you are leading this because this is sometimes going to be led by CEOs very forward thinking CEOs are thinking about AI and what it means. So, in those cases, marketing managers are going to have a slightly easier job because they've already got someone high up in the business that's thinking about this. But there's going to be cases where the business is scared and the business is resistant to AI and how that gets integrated. And some of that comes from personal fear. Some of it comes from fear of the business. But the marketing leader role and responsibility within that business is to make sure that they are staying on top of what is changing in terms of where their traffic, their leads and ultimately the revenue of the business is coming from and I think they would be doing the business a disservice to not be having these conversations internally. Well, if you do want to stay on top of everything that's changing in the world of search marketing, whether that's AI search or SEO or through paid channels as well, head over to exposioninja.com/mpodcast. We'll be able to catch up with our weekly um episodes that cover everything, the latest trends and also the latest strategies that are working for our global, national, international clients as well in the B2B and B2C space. in particular with like lead generation. If you are dependent on lead generation and you want to kind of get leads via these different um types of search, do head on over there and we'll share those as well. But if you want to develop the best kind of strategies for your business for 2026. This entire series is available at exposure.ninja/strategies Ninja/st strategies 2026 where we cover AI search, SEO, paid media, specifically B2B marketing, B2C marketing, enterprise level marketing. We've got you well and truly covered. So that's exposure.ninja/strategies 2026. And next week we'll be back for more where we'll be talking about how to create an SEO strategy for 2026 as well. Thanks so much for joining us and we'll see you then. Bye. Thanks everyone.