How to get your site cited by ChatGPT, Claude and Perplexity
AI assistants cite pages they can fetch, parse and quote. What decides whether your site is one of them: crawler access, clear answers, structure, plain HTML.
When ChatGPT, Claude or Perplexity answers with sources, they are citing pages their tools could fetch and quote. No one outside those companies knows exactly how sources are chosen, and nobody can promise you a citation. What you can control is whether your pages are eligible and easy to quote.
1. Let the right crawlers in
Each assistant has crawlers for different jobs: some collect training data, some build a search index, some fetch a page when a user asks. Blocking all of them with one rule, or letting a CDN's bot protection do it silently, keeps you out of answers. Name them in robots.txt and decide separately about training. The list and the reasoning are in robots.txt for AI crawlers.
2. Serve real HTML
Many AI fetchers do not run JavaScript. A page that renders its text in the browser can arrive empty. Prerendered or server-rendered pages are readable by everything.
3. Put the answer first
Assistants quote passages that answer a question on their own. A post that states the answer in its first paragraph, then explains, is easier to quote than one that builds up to it. Short FAQ answers under clear questions work the same way.
4. Be specific and first-hand
Generic advice exists on a thousand pages. A number you measured, a mistake you made and fixed, a setting you tested: that is the sentence worth citing, because no one else has it.
5. Give models a map
An llms.txt at your root lists your important pages in plain markdown, with a line on each. It is cheap to keep current if a script writes it. See what llms.txt is.
6. Measure what you can
Visits from assistants often arrive with a referrer or a utm_source you can see in analytics. We set that up in tracking traffic from ChatGPT and Claude.
SEO Blog Writer does the parts that live in your project: it writes posts with the answer at the top and a visible FAQ, names ChatGPT's, Claude's, Perplexity's and Gemini's crawlers in your robots.txt after you choose about training, and keeps an llms.txt listing every post.
Questions
Can I pay or submit to be cited by ChatGPT?
No. There is no submission form. Assistants that search the web cite pages their search and fetch steps can reach and use, so the work is making your pages reachable and quotable.
Does blocking AI crawlers stop citations?
Blocking an assistant's search or user-triggered fetch crawler can keep your pages out of its answers. Blocking only training crawlers is a separate choice. Check which crawler does which job before blocking any.
Does llms.txt make AI assistants cite me?
Nothing guarantees a citation. An llms.txt gives language models a clean map of your site, which some tools read. It helps, and it costs little, but it is not a switch.