ChatGPT SEO: what actually decides whether you get named
Nearly five times more people ask an AI engine this question than type it into Google. Here is what the work really consists of, and the three things sold as shortcuts that are not.
Start with the thing nobody checks
Before anything else, confirm the AI crawlers can fetch the pages that matter. GPTBot, ClaudeBot, PerplexityBot and Google Extended all have to be allowed through at the server and the CDN. A store can be silently blocking every one of them and have no idea. This is the highest certainty lever in the whole discipline and it is usually the one left undone.
It is still SEO underneath
The same things that decide whether Google ranks you decide whether an AI engine cites you: reachable pages, real authority, and content that says something only you can say. What AI search adds is a measurement problem, not a new set of tricks.
Measure before you touch anything
Build a fixed panel of the questions your buyers actually ask. Run it across the engines. Record three separate things: whether you are mentioned, whether you are cited, and who gets named instead of you. Most tools collapse those into one score, and the score is quietly wrong because being mentioned and being recommended are not the same event.
Three things I will not sell you
A text file that no major engine has committed to using for ranking or citation. A special schema block, because there is no separate AI version of markup that buys a citation. And the claim that ranking first wins the answer, because AI answers pull from a wider and different set of sources than the top of the blue links.
Real numbers from a live client account. Every result page carries the before, the after, and what the figure does not prove.