How Are You Optimizing for ChatGPT, Gemini & AI Search?

[MSaeed196]

I would focus more on User Intent, topical authority, strong internal linking, and genuinely useful answers For User Because Google is also saying The content is Valuable for user. Schema helps too, but I would not rely on it alone.

 

[KoD-]

In my view, i'd focus less on ai optimizing and more on making site to understand and cite. Here my priorities be strong topical coverage, clear enitity signal, first hand info, structure content and solid internal linking. Schema helps but it not treated as shortcut for ai visibility.

Exactly, AI visibility starts with strong topical coverage and clear, trustworthy information.

I've been testing it a bit differently. Instead of asking whether a page is "AI optimized", I ask ChatGPT/Gemini the same questions customers actually ask us and see who they cite.

On a registrar site those can be very specific things like transfer rules, ccTLD requirements, renewal pricing, etc. If the answer is correct but they're citing someone else for it, that's usually the page I go back and work on.

Real customer questions make it much easier to spot which pages need improvement.

 

[finn99]

Don't AIs tend to search on specific sites/directories known to them nowadays, instead of scrape the whole web like Google does?

For example, I'm writing a simple plugin that makes it easy for you to use my product on WordPress, but the value of its code is, or will be, practically null. By releasing it on GitHub, I might be able to cover one of OpenAIs sources or information.

This is more or less my current plan though. I will read @Starblazer 's guide when I free my backlog

 

[johnyow416]

only 1% internet traffic is from AI bots. why worry about AI? its media hype just like media political propaganda.

 

[ploxtanio]

That’s the upside of platforms like Reddit & Quora, they’re high auth platforms and pretty much all language models grab the majority of their information from there.

 

[SEO Sparton]

I would take a similar approach of emphasizing entities, relevancy of topics, and search intent and improve the content with proper structuring, internal linking, and schema markup.

 

[finn99]

only 1% internet traffic is from AI bots. why worry about AI? its media hype just like media political propaganda.

For now. I am not a particularly early adopter of anything. If something has been working fine for me, I tend to keep it, and only after a hell lot of time I try out alternatives that I have seen mentioned several times, etc.

Using AI to retrieve information, or a rank of products, seemed natural to me though, and now I barely go down the fold at all when I Google something. Google was also fast to adopt AI search results for a reason. I think this is just the beginning of a new search era.

 

[CampData]

Any guide on how to find actual fan-out queries?

 

[KoD-]

On-page structure and schema are definitely key but LLMs also rely heavily on off-page brand citations across unlinked third party mentions and Reddit/forum discussions. If the entity consensus isnt established across web sources even perfect schema wont get you cited regularly.

I would focus more on User Intent, topical authority, strong internal linking, and genuinely useful answers For User Because Google is also saying The content is Valuable for user. Schema helps too, but I would not rely on it alone.

of course on page optimization helps a lot

Don't AIs tend to search on specific sites/directories known to them nowadays, instead of scrape the whole web like Google does?

For example, I'm writing a simple plugin that makes it easy for you to use my product on WordPress, but the value of its code is, or will be, practically null. By releasing it on GitHub, I might be able to cover one of OpenAIs sources or information.

This is more or less my current plan though. I will read @Starblazer 's guide when I free my backlog

simple resource can discover AI results

only 1% internet traffic is from AI bots. why worry about AI? its media hype just like media political propaganda.

Hmm I dont feel like that , People give attention to AI results and they crosss check with SERP results

That’s the upside of platforms like Reddit & Quora, they’re high auth platforms and pretty much all language models grab the majority of their information from there.

Community mentions give chances to show up in AI places

 

[recyclable me]

I’d also focus on covering the topic fully and making the content genuinely useful instead of adding schema everywhere. Internal linking and updating weak sections can help too.

 

[KoD-]

I would take a similar approach of emphasizing entities, relevancy of topics, and search intent and improve the content with proper structuring, internal linking, and schema markup.

thanks for align with me

For now. I am not a particularly early adopter of anything. If something has been working fine for me, I tend to keep it, and only after a hell lot of time I try out alternatives that I have seen mentioned several times, etc.

Using AI to retrieve information, or a rank of products, seemed natural to me though, and now I barely go down the fold at all when I Google something. Google was also fast to adopt AI search results for a reason. I think this is just the beginning of a new search era.

Yeah, I’m pretty much the same. Once you get used to getting the answer directly from AI, you don’t really feel like scrolling through pages anymore. Definitely feels like search is heading in a whole new direction.

Any guide on how to find actual fan-out queries?

Yeah, I usually start with real customer questions, PAA and related searches, then use ChatGPT/Gemini to uncover the follow-up queries people are actually asking.

 

[Zytech SEO]

I'd focus on building topical authority, earning quality citations, and keeping content genuinely useful for users.

 

[k4ey]

For now. I am not a particularly early adopter of anything. If something has been working fine for me, I tend to keep it, and only after a hell lot of time I try out alternatives that I have seen mentioned several times, etc.

Using AI to retrieve information, or a rank of products, seemed natural to me though, and now I barely go down the fold at all when I Google something. Google was also fast to adopt AI search results for a reason. I think this is just the beginning of a new search era.

A fun observation about this type of user behavior with AI seo, because making it better has actually lost us money.
Basically we had optimised so good, that when the sceptical customer that would otherwise not buy, bought. And when they realised it's dropshipping from china they refunded. And when running the campaign we were CONSTANTLY having this issue. We ended up literally removing it because almost all the gains we had from that funnel were just people that shouldn't be buying in the first place. The funny part is, if you optimise well enough, you will have the AI halucinate in your favor. For example it was tweaking about domestic brand even though we were literal dropshipping from china and had return address there in our politics.