The short version
A buyer opens ChatGPT and asks for the best option in your category. The model gives three to five names. You want to be one of them.
- ChatGPT names businesses it has read good things about. It repeats the web, so most of what decides the answer is not on your own site.
- Five things drive it: coverage, reviews, your site, the questions being asked, and where you stand at each buying stage.
- You cannot fix what you cannot see. Measure where you stand first, then work the gap.
Why ChatGPT names some businesses and skips others
ChatGPT does not rank a list of links. It gives an answer. To build that answer, it draws on what it has read across the web: articles, reviews, forum threads, and product pages.
So the model names the businesses the web talks about well. If third-party sources describe you clearly and often, you show up. If they say little, you are skipped, even when your product is great.
This is the core idea behind AEO. If it is new to you, start with what is AEO.
The five things that decide whether ChatGPT names you
1. Coverage in the sources ChatGPT trusts
The model leans on editorial coverage and roundup articles in your category. A single strong piece in a publication it reads can put you in the answer. Earning that coverage is slow work, but it lasts.
2. Review volume where ChatGPT looks
Reviews on Amazon, Trustpilot, and Reddit teach the model that real people use and trust you. Volume and recency both count. A steady stream of reviews is more durable than one viral month.
3. Your own site, written for a machine to read
Your product pages and structured data still matter. They help ChatGPT understand what you sell and who it is for. Make the language plain, and match how you want to be described.
4. The exact questions being asked
The same business can show up for one question and vanish for another. “Best budget option” and “most durable option” pull different names. You need to know which questions your buyers actually ask.
5. Where you stand at each buying stage
A buyer learning a category asks different questions than one comparing two finalists. You can be strong late and invisible early. One blended score hides that, so break it out by stage.
How to measure where you stand
You cannot improve an answer you have never seen. So the first step is to read what ChatGPT says about you today.
Run the real questions your buyers ask. Run them across ChatGPT, Gemini, and Claude, with live web search on. Count only the times AI names you on its own. Note who it names instead, and where it got its information. That map tells you which of the five things to fix first.
That is what an Adacity report does. See how your report works for the seven things we measure, or read how we check whether AI cites your brand.
Start with the answer you have now
You do not have to guess whether ChatGPT names you. Paste your website and see. From there, the five levers above give you a clear order of work.
Want to see where you stand? Run the free check. You get the questions AI gets about you, the brands it names instead, and what to fix, in your inbox.