How to Get Started with Generative Engine Optimization (Ultimate LLM Optimization Guide)
Video transcript, 12:38
In this video Tanya Van Gastel explains generative engine optimization from the ground up. She looks at which AI tools actually send traffic, why ChatGPT gives a different answer every time you ask, and how people phrase questions to a chatbot compared with a search box. She ends with a small audit you can run yourself in three prompts.
Key points
- ChatGPT is the AI tool that matters most for traffic, with Perplexity far behind; Google AI Overviews appear in roughly 10% of searches.
- ChatGPT rarely sends a link, so most of its influence shows up in analytics as branded or direct traffic, not as referrals.
- LLMs are probabilistic: run the same prompt several times before drawing conclusions about your visibility.
- People ask AI very specific, conversational questions, so a page that spells out every use case and budget helps the model recommend you.
- Write your own comparison pages so the AI does not have to guess how you differ from competitors.
- Best-of lists, direct comparisons, clear product pages, original data and FAQ content are the formats AI answers cite most.
- Audit yourself with three prompts: a use-case question, a best-product question, and what people say about your brand, then check the sources.
Why AI search is a new channel
Let's talk about AI search. Everybody is talking about AI search these days because it is kind of the new frontier of digital marketing.
It is this whole new channel that has opened up in the last 2 years, and the reason for that is everybody's using ChatGPT, everybody is using AI for all different kinds of things, and a lot of the time now when we're making things like purchase decisions, we'll actually go and ask ChatGPT, hey, what do you think about this, or can you recommend me a product for this specific use case?
And so, as business owners, we want our businesses to, of course, show up as the solution here and make sure that ChatGPT is recommending our business and our solution. That's really important because everybody is now using AI to shop. This University of Virginia study said that nearly 60% of users are using it when they are looking for something to buy, and what's even crazier is that people are actually preferring this over feedback from a friend.
They trust AI more because they feel like AI understands them better. There's this huge report that's been built up where ChatGPT has so much of your information, it understands your needs, and when it comes to making purchase decisions, that's something that really, really matters.
Who Tanya is and why she works on this
So, why should you listen to me? My name is Tanya Van Gastel. 2 years ago, I founded an AI photography business that we built from zero to $350,000 revenue with absolutely no outside funding. We were able to sell that this year, and now I have started a new business which is called Rankingonai.com, where I get to combine my love for AI and SEO into helping clients get recommended by ChatGPT.
We've been super lucky to work with awesome clients, companies mostly tech SaaS, doing anywhere from 2 million to 2 billion dollars in annual revenue, and so in the last couple of months, I've spent over 500 hours in the trenches of AI SEO to really figure everything out because the field is evolving very quickly. And so, I'm going to share a lot of my learnings with you today. So, let's dive in.
Which LLMs actually matter
What is the first thing that you need to know about AI SEO? I'm going to share you five things, and then at the end, I'll give you another tip so you can best use all of this. The first question here is which LLMs actually matter? There's Perplexity, there is Copilot, there is so many other ones. Actually, there's only a couple that you should really focus on. Basically, ChatGPT is the dominant LLM driving 0.21% of all website traffic.
Perplexity is a lot smaller, but they come in second. They are at 0.02% of all traffic. That's 10 times as little, so it doesn't really matter. ChatGPT is really the chatbot that matters the most, and then there's also Google's AI overview. It happens in about 10% of queries. So, let's say that I Google for what is the best AI note taker.
You're going to see here that this AI overview, that's what it's called, it's basically where Google is combining a lot of its top first ranked searches into a generated snippet that you can then see here. We're going to talk later about what these are. They're called AI citations, but this is really the thing that matters. So, this is not someone just writing all of this out, that is a generated text, and they're called Google AI overview.
So, really the only things that matter here are ChatGPT and Google. You'll also see here in this traffic sources comparison that really Google is still driving most of the traffic. It's still driving like 40% of traffic as opposed to ChatGPT driving 0.21%. So, even when you go and look in your analytics, actually you have very little traffic from GPT, but that's because of two reasons. ChatGPT doesn't generally send you a link.
So, a lot of the time, people are seeing this recommendation, maybe if you ask for the best AI note taker, GPT says Fireflies is the best, but it doesn't give you a link. So, then you have to go in and Google for that company, and then you get on the link. So, what does that mean?
Basically, in your Google, in your website analytics in the back end, it doesn't say that your traffic is directly coming from ChatGPT, and that is something really to be very mindful of and keep in mind throughout your entire GEO journey because tracking becomes very difficult. But there are some ways that we can do that, and we can talk about that as well. That's first.
LLMs are probabilistic
Second, ChatGPT and other LLMs are probabilistic. What does that mean? That's a complicated term to basically say you will get different answers every single time. So, when you're doing your audit, when you're figuring out how popular your company, your solution actually is, you're going to need to ask the same question multiple times. You're going to get different answers every single time. That is a key feature of a lot of AIs is that every answer is unique.
So, try the same prompt two times. I'll give you an example. Again, we're going to ask, what is the best AI note taker? And I'm asking this to ChatGPT this time, not in an AI overview. So, here it's telling me that Otter, Fireflies, Jamie, MeetGeek, all of these. Okay, great. Let me ask another time. I'm just going to copy that, and I'm going to put it in a new chat.
Remember, it gave me Otter as number one the first time, and this time, it gave me Jamie as number one. It even recommended a couple of new ones that weren't there before. Sonnet, Superpowered, Tactiq AI. And so, here you see so clearly, every time AI will give you completely different answer, so it's not consistent at all.
It is probabilistic, and that is something you need to keep in mind because sometimes you will be recommended, and sometimes you won't, and there will be patterns for that. You basically want to go and figure out what those patterns are, but you really want to make sure that you use your prompt couple of times before you go and make conclusions about your actual AI visibility.
Conversational queries change your pages
Third, people are using conversational queries when it comes to ChatGPT. Especially when you're coming from an SEO background, you're thinking, oh, I can see search volume in things like Ahrefs or Semrush. You can basically with SEO see how many people are searching for a specific word on Google because Google shares that data. ChatGPT doesn't share any of that data. And more importantly, ChatGPT actually, you're asking very specific questions.
So, instead of asking, what is the best AI note taker, you will ask things like, what is the best AI note taker that also does video? What is the best AI note taker for less than $100 a month? What is a free AI note taker? What is the most premium AI note taker that also works for teams?
So, very specific kind of questions, and that completely changes how you think about your feature page, or how you think about how you are offering solutions to users on your website because when people are asking such specific questions, that means that ChatGPT needs to understand in a lot of detail what your company actually does, what the specific use cases are.
And a lot of the time, unless you've written them out very specifically on an about us page, it's going to be very difficult for you to exactly quantify or show the kind of use cases that you have. So, a huge tip that I recommend to all of my clients is make a page with all the specific use cases that ChatGPT can go and learn what your specific use cases are. Is it under a specific kind of budget?
Is it above a specific kind of budget? What are all the features? What are all the combinations of features? And so, the more in-depth that you go there, and if that page is fed into ChatGPT in like your audit, the more specific that you go, the more likely that ChatGPT will recommend you. So, it's really important that you know in these conversational queries, ChatGPT will almost always recommend a couple of different options.
So, you are much more in competition with other companies.
Write your own comparisons
And so, one trick or one solution to overcoming this in your content is to do your own comparisons. In your own blog page, write comparisons of you and your competitor so that ChatGPT will actually go and use that as a reference.
And so, it's very important that in these competitor pages, that you actually highlight your strengths because otherwise you're counting on AIs like ChatGPT to actually make that comparison for themselves, and they may not always find out all of the features that you have. So, go and make a lot of comparison content, and it will be hugely beneficial for your AI search strategy.
Content formats that get cited
Fourth, you really want to focus on creating content that people ask questions about. So, this goes a little bit into the conversational queries, but you need to focus on high citation content formats. What does high citation mean? Citation basically means which things are ChatGPT or Google AI overview referencing? So, if I ask, for example, right? I ask for like the best AI note taker, you can see it give me a bunch of recommendations.
When I'm asking, who are the best AI note takers, it doesn't know that automatically. It's not in its training data.
It actually went to go and search for all of that, and studies have shown that ChatGPT is actually using Google's top ranking content to do a lot of these searches, and here you can see it's coming from Zapier who wrote best note taking apps listicle, another best note taking apps listicle, a YouTube video about it, some Reddit, and even Jamie's own AI note taker app listicles. So, let's actually go and see this article that they wrote.
They wrote AI note takers, we tried the best seven in 2025. Of course, they actually put themselves as number one, as they should. And you can see they wrote a really thorough article. They even have this image about best free AI note takers. That's really smart because it helps you a lot for rankings to have really good comparison images. But I made a ultimate GEO playbook.
This is basically a complete strategy from start to finish with everything you need to know to rank on AI from your like ability audit that you can do, your technical setup for non-technical people. I even templates for how you can ask your webmaster to make sure that you are in things like Bing webmaster which feeds a lot of ChatGPT data. So, I thought of it as like, this is everything that I've learned about GEO in this playbook.
I also have a lot of these image comparison templates in Canva so you can easily swap in your images and then use that for your competitor comparison content. I also some of that content, content tweaks, everything really. So, and you can go and check that out in the description below. Again, create content that people answer questions about. I've already mentioned these best lists work really, really well. Other ones that work really well are comparisons, direct comparisons.
So, Jamie AI might make one that is just like Jamie AI versus Fireflies, product pages with super clear features. All those use cases work really well, too. Data, authoritative research, and then also FAQ content. FAQ content works really well because basically with conversational queries, users are asking ChatGPT a certain question. If you have already answered that question exactly on some of your pages, then ChatGPT is very likely to go and take that information.
It will always paraphrase it, but that's really something that's very important to know.
A three-prompt AI visibility audit
Lastly, I wanted to share an actionable takeaway with you. So, now you have all of the information, what are you going to do with it? That you go and do a tiny little AI visibility audit. So, I'm going to leave you with three prompts to see if you actually show up on ChatGPT. Think about this as an exercise also about how well you understand your company and how you want to be talking to your consumers.
You can just go in, copy and paste these. I'll put a link to this LinkedIn post down below so you can easily do that. But really this is about figuring out where you stand right now so after you take all of that action, you have a comparison. Here's an example. You need to have a solution for a specific use case.
So, let's say again you're an AI note taker, you want to go in ChatGPT and say I need a AI note taker for teams. What should I use? Then see, see what it's recommending, see if it recommends you. What is the best product? Again, what is the best AI note taker? You're going to get different kind of solutions for this from ChatGPT. It's going to give you different answers.
And here it's really good because it will usually break it down in a couple of different things and it will compare you as this product is maybe better this, this product is better for that. Go fact check that, see if you agree with it. This is a really big one. It's what are people saying about your brand? So, ask it specifically what people are saying about your company.
You might learn some things that you don't like, you might learn some things that you really like, but it's really important to know. And then also go and look at the citations, the sources like I showed you earlier because that's really going to show you where all of that information is coming from. I hope that this was helpful for you guys. I'm going to make a lot more videos about this content.
So, if you have any questions, please ask them below and maybe they'll pop up in the next video or so.
Transcript lightly edited for readability: filler removed, punctuation and names fixed, her wording kept. All video transcripts.
