HTTP Status Code
An HTTP status code is a three-digit number a server sends back when someone requests a page. It says, in shorthand, how the request went: 200 means "all good", 301 "moved permanently", 404 "not found", 500 "server error". AI crawlers receive this code too, and it helps decide whether your content gets read at all.
Why status codes matter for your AI visibility
AI systems like ChatGPT, Perplexity or Google AI Overviews gather their knowledge by having their crawlers fetch pages. If such a crawler gets an error code instead of a 200 when it makes the request, the page often isn't read at all or is judged unreliable. A page with strong content that technically responds with a 404 or 500 simply doesn't exist for the AI. Status codes are therefore the entry ticket: only what is delivered cleanly with a 200 can be cited, recommended or built into an answer. Before you polish text and structure, this foundation has to be right, meaning the correct code at the correct address. Otherwise the best optimization fizzles out, because the machine never arrives.
How status codes work
Every request for a web page runs as a question-and-answer game: the browser or crawler asks for an address, and the server responds with content plus a status code. The first digit reveals the category. 2xx means success, 3xx means redirect, 4xx means an error on the requester's side (a wrong address, for example), 5xx means an error on the server side. The ones you'll meet most often are 200 (success), 301 (moved permanently, address redirected), 302 (moved temporarily), 404 (page not found) and 503 (server currently unavailable). The code sits in the invisible header of the response and can be read at any time with the browser's developer tools or free online checkers.
Common mistakes
A typical stumbling block is soft 404s: the page shows an error message but still reports code 200 to the crawler. That way empty or wrong information ends up in the AI index. Just as harmful are redirect chains, where an address goes through several 301 hops until crawlers give up in frustration. Confusing 301 (permanent) and 302 (temporary) also costs visibility, because only with a 301 does the reputation built up by the old address carry over. And a temporary 503 that lasts too long can cause important pages to drop out of the index. Check your most important pages regularly for the code they actually deliver.
Relation to AI recommendations
For an AI to name your brand or link to it as a source, it first has to fetch your page without errors. Status codes are the technical prerequisite that sits above everything else: the perfect answer page with clean structure and strong citability is useless if the server refuses it with a 404 or 500. For good AI visibility the rule is therefore: target pages return 200, outdated addresses redirect via a clean 301 to the new version, and error pages are rare and short-lived. That way you ensure that AI crawlers like GPTBot can take in your content in full and factor it into their answers.
Example
Imagine a trade business that moves its popular guide page "Cleaning your gutters" to a new address. If it forgets the redirect, the old address now returns a 404. ChatGPT and Perplexity had cited this page before but can no longer find it, so they drop the recommendation. If the business instead sets up a 301 from the old address to the new one, crawlers follow the move automatically, and the visibility it built up stays intact. A single status code decides here between staying and disappearing.
Common questions
Which status code is best for AI visibility?
Code 200 for every page that should be found. It signals a successful request, so AI crawlers can read the content in full and reuse it. For moved pages, use a 301 redirect.
What is the difference between 301 and 302?
301 means moved permanently: the reputation built up carries over to the new address. 302 means only temporary, which is why search engines and AI keep the old address in the index. For genuine moves, always use 301.