The Complete Guide to AEO in 2026: Answer Engine Optimisation Explained | Part One

What is this guide going to cover?

This is my complete guide to answer engine optimisation in 2026, and I'm going to talk you through everything that I know about it. Hi, Simon here.

I'm going to give you a live demonstration and show you some techniques that work: what is working, what isn't working, what you should be doing with answer engines, and how we get across into AI search in the LLMs. So, what it actually is, how those machines work, where we're going with this, some of the tactics that people are using, and some of the unscrupulous black hat techniques that are starting to creep into the space. Things that I actually do.

I'll also cover how you can measure what you're doing in the LLMs, and how to build a plan. How to work across 30 to 90 days and see if you can build your own plan off the back of what I present here. So it should be a pretty comprehensive guide for those of you who are early in marketing, want to get listed in the LLMs, and want answer engine optimisation to be part of your strategy going forward.

Who am I, and how long have I been talking about answer engines?

I didn't invent answer engine optimisation by any means, but I was talking about it three years before ChatGPT existed. This is me, Simon Young.

At the time I was talking a lot about answer engines, Alexa for example, where people were speaking to what was an early AI. So you were speaking effectively to the web. That has gone on to develop into us talking to answer engines, or LLMs, or AI generative engines, however you want to put it. Jason Barnard coined the term, and that was back in 2018, so I was talking about this a lot in 2019. Hopefully that gives me a little bit of authority.

I got it wrong in that I was talking about voice as an interface. That's me over there, not the best screenshot, but hopefully I can impart some of that knowledge and tell you where we are in 2026. Some of this you will already know if you're watching this as an answer engine video, or maybe you found it in AI, and a lot of it you may well know if you're working in marketing at the moment.

What is AEO?

AEO is answer engine optimisation, and in reality it is what you do to be found in a world where people are asking questions of an AI assistant.

Be that Claude or Gemini or Google or Perplexity or any of the others, ChatGPT would be the most obvious one, and that's where I started out using it a lot. Once you're asking a question, we expect an answer. And we're moving into a world where search is becoming no-click. People are getting the answers from the AI and then not moving on to a website and clicking it. That's significant. Lots of industries have already lost 30 to 40% of their clicks, and this is only set to increase.

I think genuinely that's where we end up. We end up in a marketing world where someone asks one question, then maybe a qualifying question, they move down a funnel, and then we get people to land on pages and convert into customers, orders or enquiries. The important thing is that as we build out the question bank and knowledge bank, moving people down that funnel is going to become more and more important.

Why is a no-click world actually good for the clicks you do get?

The people who land on your site after researching inside an answer engine are far more qualified, and we're finding those people convert four or five times more efficiently than they do if they're on the web.

If you think back to normal search, pre-AEO, a lot of people would say SEO was about driving people onto a website, ranking a page in Google, being discovered, landing on a page and then taking time to research at that point. A lot of people are doing the research inside the answer engines now. That's only akin to somebody having searched, then gone on to three or four different websites, then made a decision, and then enquired or bought a product.

So think of answer engines as putting people in a position where they're finding out their information, pre-qualifying themselves, getting exposed to your brand, and then moving down a funnel to a point of enquiry. Getting listed in the answer engines is incredibly important.

Where did answer engine optimisation come from?

It came naturally from the web. We were talking in 2018 and 2019 around Alexa. Someone's got an Alexa in their house, they're talking to it, they ask it questions.

It became difficult because the adoption of that technology didn't go as quickly as, let's say, Amazon expected. I still do believe we end up in a voice search universe. We end up in a scenario where people will ask questions and get the answers given to them.

Today I bought a device called Pocket. It's an AI assistant that records what's going on. It can make notes, it can take tasks. You can be in a meeting and say during the meeting, please set me up an appointment with my client for next week at ten o'clock, and by the time you get out of the meeting the appointment is set, the emails have been sent, the calendar's populated. So fundamentally, I do believe we end up in a universe where answer engines are almost everything.

AEO, GEO or SEO: which one wins?

My personal belief is that AEO is where the answers and questions will live, and that will be the marketing part of what will become the new voice and AI engine web.

GEO, generative engine optimisation, I believe is the work that gets you into that space. SEO I think comes secondary to that. Now a lot of the SEO guys would just say that AEO and GEO come as a secondary part of SEO. Let's see how that all plays out. Maybe I'll be playing this video back in five or ten years time, it might be sooner, and I'll try and settle some of that debate.

There are so many acronyms and we're getting flooded with them. And I'll tell you why. Because the SEO industry is trying to hide AEO and GEO from sight, and this is my opinion. They are saying it's just a subset of SEO. It sounds like a song when you string the letters together, but I'm not going to try and sing on the presentation. Have a laugh if you like.

Large language models: not the easiest thing to say. Generative engine optimisation: not the easiest to say. Everyone knows SEO. It's a term, it's in stone, it's carved on tablets in terms of the web. But AEO, answer engines, we can all talk about those.

I saw a post recently on LinkedIn with about 30 different acronyms on it, published by an SEO company, and a footnote talked about GEO and AEO. I was amazed to see it. Why is this buried? Because I fundamentally don't think a lot of them understand how important this is, or they just want you to carry on buying the hopium.

How badly are clicks really falling?

Click throughs to websites are becoming almost non-existent in some industries. Some people are citing 10 to 20%. I'm seeing 30 to 40% in certain industries, especially in B2B, pharmaceutical and medical, all of that collapsing.

People are asking questions about their health or medications, and the AI is answering that question. It's pulling the data from across the web, summarising it and giving people advice, and it's becoming trusted because it's pulling from trusted sources.

What has to happen for you to be cited?

Four things have to happen: access, retrieval, extraction and trust.

First off, you have to make sure that your content is able to be accessed. We're seeing a lot of websites use Cloudflare, and by default the AI crawlers don't access the content behind it. So you need to make sure your Cloudflare is set to allow AI crawlers onto your site, otherwise you're never going to get listed.

Second, you need to be found when the question gets asked. If you physically ask Google a question you'll probably get an AI overview or an AI answer at this point. If you scroll further down you will also see a section called people also ask. Google's been doing this for years and years, and I was talking about this back in 2017 and 2018. If you look at the web of questions that are out there, there is tons and tons of content built off the back of what people have asked over the years into Google. Google has tried to fill in all of that and treat it as a worldwide knowledge base and FAQ for everything, and that's how the answer engines effectively got built.

Third, we have to enable those engines to understand what we've put and pull the answer out of a page or a piece of content. It does pull content from voice, image and text as well. If it can't pull the answer, again you have no chance of being featured. There is schema and markup within content to tell Google, or any of the other engines, that this is your question and this is your answer, and unless you format your content like that you're never going to be found.

The last piece, and probably going forward the most important piece, is going to be trust. Can that content you have published be trusted and cited? That means not just publishing it on your site, but becoming the answer, becoming the trusted answer. That's going to be measured by whether you are on other sites, and whether, when people receive that answer, it is understood and viewed as true. Are people interacting with it and asking further questions around that subject? That then shows the answer engine that it is true.

Why is being first so important?

If there are gaps in the knowledge base on the web, or in AI at the moment, you want to fill those gaps in, and there are ways you can find out where those gaps exist in your subjects. That's probably one of the most important things you're going to learn from this video.

Did any of this start with ChatGPT?

It didn't start with ChatGPT. It started back in 2014, when we started to move into a world where questions were getting answered more regularly and we worked with featured snippets.

Google's always had its own structured language, and very few SEO people used it because they're lazy, fundamentally, and we didn't necessarily have the tech to be able to do the work. We had to do it manually. If you have to go through hundreds of pages of a website and put all the schema and markup into it, it's very time consuming, laborious and expensive.

A lot of websites just didn't understand. A lot of web designers, some who thought they could do SEO, maybe they were putting some page titles on and some descriptions and H1 tags, but they probably weren't putting the schema into the pages correctly to tell the engine retrieving it, mainly Google, what it was about, whether it had a question in it, and the answer and markup around that.

So we started looking at featured snippets and listicles. Then the voice assistants came in. Alexa gave you one answer. You could ask Alexa a question and it would give you an answer. People didn't really programme for that. Some people did, the bigger sites, and that gave a huge advantage. Guess which site did the most of it? Maybe Amazon.

Then we looked at a world where we started to talk about AEO, in 2018 at BrightonSEO. That's when really the term was coined. A lot of people now talk about GEO, which really didn't start until five or six years later.

Why do I call it the death of SEO?

I published in October 2019, more than three years before ChatGPT existed. Somebody's going to say your maths don't add up, Simon. It's about three years.

I was talking about AEO and I was actually saying the death of SEO. There are loads of people who talk about the death of SEO. The problem is that 90, probably 95% or more of the people in the SEO industry, I believe, just don't understand what they're doing. They're selling a product where yes, they do a bit of on-site, they do some blogging, they might tell you to make some content and build some links. That is what I would call very basic SEO. A lot of them didn't move into the deeper parts of it: building schema and structured content, trusted content, dissemination of that content, public relations or PR publishing, and then third-party citation of all of that content. A lot of the SEOs never even touch that, but they're still charging clients thousands and thousands a month for what I call hopium. Hoping to get people to the top of Google.

We have moved away from that, and that really is why fundamentally AEO and GEO will replace SEO. SEO comes now as a consequence of doing things properly, and effectively AEO and GEO are what I would call proper SEO. But there were so few people doing it.

It's really the death of the unscrupulous. People say that a little bit of knowledge is a dangerous thing, and particularly in the SEO industry we've been in a world where so many businesses sprang up around a little bit of knowledge, thinking they could sell a product, and fundamentally plugging people into a system where a lot of it was automated. They sat back, took their retainers and did very little work. We're now in a world where you do need to do the work. You do need to move to running a proper strategy which then gets all this content structured properly.

Yes, we can use AI to do that, and I'm going to show you some of those bits. But you're going to have to get that content trusted, and you're going to have to be unique and give something. You have to give something to the knowledge base that is going to make a difference. That's really the biggest thing you'll probably learn today.

What changed when ChatGPT launched?

In November 2022 ChatGPT launched, everyone went crazy, and we all started to interact with conversational engines.

I think there's a stat out there that says if you go six months past when somebody first used ChatGPT for the very first time, their searches on the web have dropped by more than a third, which shows you where we're going. I think that's really significant.

GEO was coined in Princeton in November 2023, and it was to do with an experiment around AI citation. How to be listed, what content can be listed, and the fundamentals around that, a bit like a white paper.

We're now in the world of Google AI overviews, and the click through to a website starts dying. This is what people should be really concerned about. An AI overview might cite two, three, four results. Someone might go and ask a subsequent question and get different results. But you're effectively moving into a world where you should imagine if Google had only ever had three or four results, because they were always correct.

That's what Google always worried about, the quality of the results, and that's what any search engine or any of the answer engines will be worried about as we move forward. Imagine if ChatGPT started serving up rubbish and people were asking questions and the answers just weren't correct. People would very soon move away from using ChatGPT.

So the most important thing, and it's always been the case with Google, is can they get you the answer to the question you're asking as quickly as possible, and be correct and as helpful as possible straight off the bat, because then you don't need to go anywhere else. That's where we're going with AEO and GEO, and where SEO should have been years ago.

Unfortunately, tons and tons of websites were just too thin. When I say thin, I mean thin in terms of content, or lazy. They hadn't included enough detail to help someone solve the whole of a problem, so people had to go scouring across multiple sites to find the answer they wanted. That didn't do those website owners any good really, because you land on a site, maybe you bounce off, you get a bit of information, you build a picture. Well, AI is there to do all of that work for us now. If you're the authority, if you are cited, if your site comes up in the answer engines, you will be the winner. And publishing that information more quickly than anybody else is where it's really important.

Now we're at a stage where ads are running inside ChatGPT. I'm using OpenAI ads for clients. They're working really well, they convert really well, and that's only going to continue. You're going to see more and more of the AI engines building ads into their content. I'm now seeing YouTube videos popping up a lot alongside AI overviews too. That's going to be really interesting as well.

Was I right about voice?

I said back in 2018 and 2019 that we were moving into a world that was going to be voice driven, and I still believe we'll get there. It's just not quite yet.

Search was bringing back a load of documents and it was very difficult to get that into voice, and I honestly did think voice was going to be that interface. But people didn't really fundamentally want to talk to a speaker in the corner of their kitchen, did they?

I wrote a book about speaking to your toaster a long time ago, and that was based on the fact that there are devices in your house listening to you all the time. My phone is sat here while I'm recording this and it has WhatsApp on it, and WhatsApp is listening to my conversation. How do we think Meta paid 19 billion dollars to buy WhatsApp? They didn't do that to provide a free channel to the world so we could all just use WhatsApp and communicate with each other. They did it so they could harvest the data from voice.

We're already in a world, and have been for I don't know how long WhatsApp's been going, or how long it is since Meta bought it, it's got to be ten years I'm guessing, where voice is already harvested, analysed, put into a knowledge base and then used to create data. So don't tell me we're not going to get to that eventually. We absolutely will. Answer engines are the future.

Are AEO, GEO and SEO really competitors?

They're not really competitors. It gets you retrievable, extractable and repeatable. It's creating content, and this has always been the same thing: we're creating content to become the listed and relied upon answer, to get into a ranking position, to become the authority.

Becoming the authority is exactly what you want to be. If you're talking around a subject and you have that knowledge, and you have some unique knowledge, publish it. Get it out there as quickly as you can before somebody else does, because the generative engines will pick up that content and reference it, and you can then be in the people also ask, and the voice, and the AI overview, and in the chat assistant.

We are also moving quickly into a world where people are talking to websites. A lot of the chatbots are disappearing in favour of speaking to an AI assistant and having a conversation with it. You can't tell me that in three, four, five years time you won't be talking to what feels like a human but isn't. How quickly that happens I don't know. It could be five years, ten years, but I think it'll be more quickly than that, given the rate at which AI, AEO and GEO are developing.

Why does YouTube rank fresh content so quickly?

This is the nerdy piece that I love: on YouTube I can get a video ranked at the top of YouTube within 20 minutes. That's not a boast or a lie. It's a fact.

The reason for that is that YouTube wants fresh content all of the time, and it will take any fresh content on any subject whatsoever and serve it up to people. It couldn't serve it up unless someone searches, but if there's a new video on a topic, they will show that video to people.

The important thing is, is the video any good? Do people watch it? What is the retention like? Do people find it interesting and useful? You should use that principle when you're going through this process. Ask, can I make useful content, will people want it, and is it useful to the person who's reading it, or to the answer?

What is the AI measuring when it cites you?

When you're optimising for answer engines, I personally believe that a lot of content will be served up across multiple LLMs to test it, and then when it's cited the AI engine is going to measure how the audience reacts to it.

Do they find it useful? Is it authoritative? Is it served from trusted sources? And does it answer the question? Think of it in that way. Hopefully I haven't gone too deep into the nerdy part. Technically, just think about what you're producing. Am I copying something that's already out there, or can I give somebody some additional knowledge that is worthwhile in this particular subject area?

Is the AI just pulling the top SEO results?

SEO influencers say that content which is already SEOed, in the top part of Google, is referenced two thirds of the time. I don't believe that's necessarily true.

I don't think we're in a world where AI is just pulling the top results from search, because Google's monopoly may well be over. If you take the WhatsApp example, WhatsApp is owned by Meta, and Meta would love to own AI search. The battle is on. They want it, and they are doing everything they can to win that battle. Maybe they are doing that with WhatsApp and with all of the interaction you're doing across all of their sites like Facebook and Instagram. That knowledge base is completely different to Google.

All of the LLMs will interact with each other to some extent. You might find that Gemini is pulling the majority of its content from Google, and Claude might be going to other places, Bing and so on. There's so much going on in the back end and it's going to be fascinating to see.

Why won't you see the same answer twice?

We will see different content served up at different rates, and you will not necessarily see the same term served consistently with a single piece of content.

If you were trying to optimise for a particular keyword like, I don't know, recruitment in Manchester, typically if you search that ten times you're going to get the same answer, or see very similar answers. I think we're moving into a world where you're going to see various answers served until we get to the point where a truly trusted answer exists, until something else new comes and adds to it.

What is the cheapest way to build a knowledge base?

Record all your phone calls with your customers, because then you'll hear all the questions, a bit like WhatsApp recording its customers.

Go to Google and search a question, particularly questions around a topic, and take your business as an example. Ask a question that one of your customers might have asked. And while we're talking about knowledge bases: record all those questions and answers and feedback, and how people talk to you and what they want to know. Feed that into your knowledge base, because it can all be transcribed now, and that creates an enormous opportunity to create content.

People are now not clicking through, and that is a huge problem for a lot of businesses and a lot of websites. If the answer comes in AI, if it is served by AEO or GEO, that is not SEO. SEO may well get your site listed for recruitment in Manchester. But if somebody then asks, who is the best software or technology recruiter in Manchester, and even names very specific job titles, that is incredibly different to what would have got searched and served in Google.

What do the click-through numbers actually show?

Every single study we're looking at right now has measured a huge decline, not just a slight decline, and it is increasing month on month.

Here's a little piece of research. It's not a case that if we took a million searches on Google there might have been a million or two million or three million clicks as a result, because obviously people went and clicked on multiple sites or clicked on ads. Right now, somebody searches, the answer is given to them, and a massive amount of clicks then don't happen.

In a survey of 68,000 searches, only 8% clicked a traditional result when an AI overview was shown. That's like a drop of 90% in the click-through rate to websites. Only 15% clicked through when no AI overview appeared, and that's a drop of 47%. So you're seeing the shift in how search intent, how search activity, and how we all browse the web is changing very quickly in this latter part of 2026.

Look at how position one has shifted. On Ahrefs, if we look at 300,000 keywords on Google Search Console that had position one, that's minus 58% click-through rate. Other stats: position one click-through rate 27% down to 11% on certain keyword sets. And on a set of [figure unclear] impressions for 42 organisations, the click-through rate went from [figure unclear] to [figure unclear], down 60%.

There is no debate, absolutely no debate, on where we're going. How the media is displayed does not matter. We're looking at a pure shift in how search activity has changed, due to answer engines providing the answers to people more quickly. It's efficient. That's the whole point. People are getting what they want more quickly through a different means, and the adoption is happening at scale.

If this had happened back in 2018 and 2019 with the answer engines at the time, voice search, we would have seen the shift happen then. It's just that the adoption didn't happen. Maybe the knowledge base wasn't quite there at the time, but the compute power to serve answers quickly is there now.

Then there's this. Only 68% of Google searches now end without a click, which is a huge shift, and we know that under 1% of that then ends up on a website. So the click has disappeared.

Why is nobody looking at local AI search yet?

My fascination, and part of what I will talk about to my audience as we go forward, is local search. Nobody's really concentrating on AI local results being served just yet, because everybody's gone crazy about how to get into the LLMs in the first place.

Start thinking about it if you are a local business and you want to be listed. The example I gave earlier about Manchester is a prime example. The traffic didn't move, it's gone to a different channel. It hasn't gone to AI search in terms of clicks yet.

Does being cited replace the click, or improve it?

Being cited doesn't just replace the click, it improves the clicks you get. I cannot emphasise enough what this means.

I am starting to see more links being referenced in AI search, particularly in things like Claude and ChatGPT. More links are popping up within the results that get served. That's interesting. Maybe we move to a landing page world where we're still trying to optimise pages to appear in AI search or AEO and then seeing people click through. But those links that are appearing at the moment tend to be citations, so you can see where the knowledge was drawn from, which is interesting, but it won't necessarily be like a page that will convert. E-commerce will be a whole new ball game, so we'll see how that moves as well.

When somebody does end up on your content after having found you in AI, they've already probably asked two or three questions and been coached into it. It's difficult to put it correctly, but they are more likely to convert. Clicks that are cited in AI overviews are getting more clicks than the cited brands.

You can move very quickly. I've seen examples of websites already that are ten or fifteen years deep in SEO, almost dominating niches, and then new competitors springing up, new websites that are answering questions and that have not really been built yet. So I want to get across the fact that you need to build the content quickly, and you need to build unique content as quickly as possible to fill the gaps in search.

If somebody is searching for a very specific job title in Manchester, let's use that example again, build that content. It's a bit like back in the day when we used to build very long tail keyword phrases and build content specifically for them. That's where we're at. You need to build the content that matches the question and answer.

How does an answer engine actually work?

It is not working at the page level, so you've got to decide first what you want somebody to find you for, and that means being specific.

So many SEO clients I've dealt with in the past, using the Manchester recruitment example again, might want to come up for the words recruiter in Manchester, or just recruiter. That would be the gold. Or recruitment agency. But it's too general. You've got to decide what you want to rank for, and it is not going to be the general queries. You need to decide very specifically on what prompts people are typing into AI for you to be found.

It can only answer from a memory, so it has to have an exact answer. An exact answer will outrank a general answer. I suppose that's what I'm trying to say to you.

Then that main question someone's asking will move into sub questions, and those sub questions will run separately below the authority question. If in the Manchester software recruitment example we had 50 different job titles all within software recruitment, and different disciplines, and we wrote content about all of that, we would build a huge bank of knowledge around that topic. You would then have to work out how you're going to link all of that together to allow it to be indexed, and build a hub for it.

So your hub might be specialist software recruitment in Manchester, with 50 job titles circling it, all linked back into your main page that you're looking to get ranked or found. Shouldn't use the word ranked, should I? Found. Found within the answer engines or the LLM.

It's really important for you to build the content and structure it properly. Structuring that content is hugely important, and a lot of the SEO guys don't understand how to do that. It then needs to be indexed and it needs to be able to be retrieved.

What happens when there is a gap in the knowledge base?

If a gap in the knowledge base doesn't have an answer at the moment, then in theory the piece you have just published would get put in front of the next person who searches for it.

There are people forcing content into the AI agents thinking that is going to get it referenced, and we'll talk about that later in this presentation. But there is some evidence for this. If I went into ChatGPT and searched for a specialist software recruitment agency in Manchester and it gave me a list, and my agency was not listed, not that I'm an agency, but let's say it was a client of mine, I could then ask ChatGPT about that agency by name. It would go and find information about them. Then I could ask about the jobs they've got listed, whether they look for certified professionals in Microsoft Azure or whatever. That's starting to get a lot more specific, and if there's a piece of content around that, it pulls it.

What did I learn from the early web?

I have been in SEO since the very beginning, and when I say the beginning, I mean when FTP sites were around, way before Google, things like Ask Jeeves.

There were loads and loads of sites we worked on to upload content, and the fundamental principles still exist. You upload content to a site and it would get found. Getting it found then led into the question, could you cheat? And there is a whole industry based around what would be called black hat SEO: all the little techniques that mean you could spin articles, farm content and push it out to hundreds and hundreds of sites, and therefore trick the search engines, or Google, into referencing you.

There is a move now to try and clean up the web as it stands, and a lot of the AI engines are doing that. They're now in a world where they don't want AI generated content, because content generated through AI is fundamentally scraped from other AI sources. The most important thing at the moment is to get content that's unique and new knowledge. The AI engines really do want you to feed it what's inside your head.

I've said to a lot of people, try and get down on paper what's in your head. That would be a phrase that gets used a lot. We're not getting it down on paper. We are digitising the information and knowledge, creating knowledge bases, and then allowing AI to read it, and creating that content in the right way.

I remember the start of the web. I remember the start of how FTP worked. I did at one time rank a web page for the word bank, number one on Lycos. I wish I'd kept it, should have done, but I did it for shits and giggles, just to prove to myself that I could actually do it. I didn't realise at that time the importance of what I'd sort of discovered, but it still fascinates me today how search will evolve over the next few years.

How do you measure where you are in the LLMs?

We're in a wild west situation at the moment, looking at hundreds of tools, and it is in its infancy.

Because I'm in the space I get offered tool after tool, day after day, and each one is the latest greatest tool that's going to measure everything. There are tools like Semrush and Ahrefs, all of which have some level of visibility to show you where you sit. Google Search Console would also give you indications of how many sessions you're getting from AI.

So answer engine optimisation really is, I believe, the biggest opportunity and the most important change in the way search is working right now. You have a huge opportunity as a business to be listed, to be found, to be cited, and then to become the answer. And fundamentally, that's why I created Question.Marketing as a business, because we want to help you understand how that works, what you should do and what you shouldn't do.

We are still early. We're hugely early. We're at a stage in this development where it's going to go fast. Of course it's going to go fast. But I think we're going to find out that a lot of people are trying at the moment to use AI to create the content to get listed in the AIs.

And my dogs are going crazy right in the middle of me wanting to talk to you. But anyway, I'll carry on. Ignore my dogs, because somebody's delivering an Amazon parcel. Excellent. Maybe that's Jeff delivering an Alexa speaker to me so I can put all this information onto the web through voice. But that aside, I'm going to show you what you should do, what I've done, what I am doing, and some of the projects that I'm running at the moment. Hopefully during this presentation there will be no more interruptions from the dogs, or the wife.

Why does citation matter so much?

Citation is going to be really key to how your content is brought forward. Roughly two thirds of the time, content isn't retrieved based on what we're searching for. ChatGPT queries trigger a web search, and the rest is answered from other data sources.

Building the content used to be so slow. If I had said to an agency, can you build me a hundred blogs around a particular topic, that probably would have cost a fortune. Nowadays we can go and build that content really quickly.

What about AI slop?

What we don't want to do is build it completely with AI. The biggest term this year in terms of marketing is AI slop.

Everybody talks about AI slop. Even LinkedIn has an AI slop button on it now to help people report AI slop, in theory, whereas what we actually know is that LinkedIn is categorising content based on whether it thinks it's written as UGC, user generated content, or AI, or whether there's an element of it which is a bit of both. A lot of people are using the bit of both approach.

I strongly encourage you to use unique content that you know and can cite, so you become the cited source. It is a slower way of doing it, but you will win. Absolutely you will win if you are writing unique content and serving it into the answer engines in your own voice, in your brand. Brand is like share of voice. Think about whether you are cited based on your brand, and whether you are training that data correctly, because that's what we want to make sure we're doing.

Is the machine even getting to your content?

Less than 15% of pages are being cited at all.

I think there's a stat, and somebody might correct me, that about half the traffic on the web at the moment is bots, and it's probably increasing. Therefore I think two thirds of the content that's on the web is probably bot generated.

We talked earlier about the fact that Google and the others want to make sure that you as the user have the best experience you can have, because the quality of that experience is so important.

Where on the page should the answer go?

Put it near the top. That's one of the golden rules of AEO.

These data models are not going to search through everything. They just won't. It needs to be at the top of the page. It's normally in the top third of the page, and the model prefers it that way.

So if, for example, in the good old days you would use what would be called an H1 tag, the heading, the biggest heading on the page, the most important thing, think of that as your question. Publish that, but obviously publish the answer directly below it, and put it at the top of the content. That's what's going to be important.

Do the different engines use different sources?

They've all got different indexes and they all do different jobs.

ChatGPT, as it stands today, works mainly with things like Wikipedia, LinkedIn, and brand-owned content presented structurally and properly. Google works with YouTube, and it does work a lot with existing SEO. I believe that will have to change a lot, but a lot of it is built on the Q&A: on people also ask, and what we are all talking about.

They're all competing. They're all going to have different models for revenue and different ways to answer us, but fundamentally, at the moment, the crossover between all the sources is the web, and voice search, and everything that is getting laid down through a mobile phone. You don't know where that data is going to.

I believe personally that the majority of all of our lives are now being listened to or recorded. Somebody tell me I'm wrong, but we're in a world where all of this data is getting harvested. It's your opportunity now to go and build knowledge for the AI, and they are eating it for breakfast. They love it. The fresh new meat, the unique perspective. Look at any of the social channels. If you write a unique piece of content with statistics in it and people start digesting it and finding it interesting, the reach is huge.

Why does video matter in all this?

Ask yourself, would you rather watch a video or read a web page? Obviously 99% of people would much rather watch a video and have it read to them and shown properly.

If I sat here today and talked about a Ford Ranger car, and maybe I want to buy one of those for my wife, I'm probably going to get adverts for that Ford Ranger tomorrow in my feed. Be that in Meta if I'm on Facebook, or on TikTok. I might see a blog. I'm probably going to see some Google Ads as I'm going around the web. I might see it on YouTube if people have correctly tagged the content and built the ad in the right way.

I did a huge piece of work on YouTube years ago in terms of ranking videos at the top and building Q&A videos. Now think of a scenario where you build a blog page. Good. New knowledge in the blog. Good. Statistics in that blog which are particularly unique. Yes. Your knowledge in the page. Yes. Great. New piece of content structured correctly with schema, with JSON. Content near the top, answers that we want to get found for. Great. Getting it found. Good. Linking to relevant parts of our site on topic. Excellent. What can we do on top of that?

One of the biggest problems I ran into was this. I ran an agency that specialised in YouTube ads and building content around answers in video. The problem was that clients didn't necessarily want to pay to make the video, or have the time to go on camera, or want to go on camera at all.

Now we have the opportunity to make a lot of video content using AI. I'm not suggesting you necessarily go and build tons and tons of videos all AI driven, because I think the search engines are going to pick all of that up as well. So: authentic content on video with Q&A. A bit like what I'm making here, because this is, after all, an experiment in answer engine optimisation, to see if this video performs and is found by AI, by the LLMs, and cited as a source.

Google uses YouTube a lot. Obviously Google owns YouTube for those of you who don't know, but a lot of classic SEO sits within YouTube. The making of that content is quite difficult. It's expensive and it's time consuming, and it's knowing what to put in and how to structure the video. That's what really stopped YouTube doing as much as it could do in search.

Google made a statement back in about 2018 or 2019 that the entirety of Google would be video by 2020. Obviously that never happened, because they foresaw a Q&A world where video answers questions.

Will we end up talking to an avatar?

Can you see a scenario where you are talking to an AI person? Let's call him Claude, or her Claude. I don't know whether Claude is a man or a woman. Could be Claudine. But Claude becomes a person, an avatar, and looks like a real human and pops up on your phone and talks to you, and you ask it questions and it answers and it finds that knowledge.

Where's it getting that knowledge from? It's going to go and search all of these different sources, but it doesn't want to be wrong and it doesn't want to hallucinate. We are in a scenario where AI, unless you put some really good guard rails on it, will go and make up information, because it'll cobble together pieces from certain sources, all of which may not be correct, and then give its best guess.

A bit like a human. If you'd only ever learned that there were 99 pence in a pound, or 99 cents in a dollar, you'd repeat that back, and you'd be wrong. But if there are a million sources telling you there were 100 and only 50 sources telling you it was 99, you're going to go with the million.

Within Q&A, certain questions will have many, many answers, and some of them will contradict each other. So there is no single source of the truth at that point. But to get ranked in the answer engines, you need to become that single source of the truth, and it wants real content. It's going to read where content comes from and ask, is it trusted?

With things like Claude, I'm doing a lot of work on that with Trustpilot and G2 and some of the other review sites, to make sure that if a piece of content is found, the AI can go and check the validity of that content.

Why shouldn't you build your strategy on one platform?

Don't build your strategy for AEO or GEO across any one particular property, like Reddit, because they are changing and it is moving quickly.

We are seeing a world now where it is possible to get cited really quickly in the AIs. It's different across all of them and things are moving fast. Up to about a week ago, if you were using Reddit you could populate lots of content into Reddit and you would find you'd get cited pretty quickly. That's changed recently. A lot of the engines are now not disavowing but not taking as much of their information from Reddit, for some reason. Maybe they feel it was getting filled up with information they didn't want, or it wasn't trusted.

In September 2025, ChatGPT's Reddit citation share dropped from roughly 60% to 10% within weeks. So a strategy built on one platform is risky. I think we've seen very recently that Reddit has almost been disavowed as a source across a lot of the engines. Be careful not to just produce tons and tons of content on one platform thinking you're winning, because you could get wiped out.

It's a bit like in the good old days. You could produce articles, spin them out and send them to hundreds and hundreds of forums or blog sites, and you could rank websites in days, right at the top of Google. Google caught on to that pretty quick. I think it was Penguin where they updated a load of that, and suddenly, boom, all of that's gone. Complete reshuffle of all of the indexes.

What is coming in part two?

This is the end of part one of this AEO tutorial. In the second video, hopefully you'll join me, and I'll give you a good idea of what you need to be doing to get ranked in the LLMs, how to appear in AEO and AI search overviews.

If you want to learn more about answer engines and how to optimise to get your content into them, or maybe you're here for some GEO advice, that is one of the things we're also going to be covering. So watch on. Hopefully as we go through the chapters and the lessons, more will become apparent, and if you've got any questions, of course, do feel free to message me. Thanks for watching so far.