GEO how-to · 6 min read
How to Get Cited by ChatGPT
To get cited by ChatGPT, a page must be easy for it to find, easy to extract a clear fact from, and safe to trust. In practice that means making the page crawlable, leading with a plain definition, marking it up with schema, writing in clear question-shaped sections, and earning mentions elsewhere on the web. ChatGPT's search and browsing modes pull citations from pages that state facts plainly and are corroborated by other sources.
1. Make sure the page is crawlable
ChatGPT can only cite pages its crawler can reach. Confirm your robots.txt does not block AI crawlers such as OAI-SearchBot and GPTBot, ensure the page returns its content in the initial HTML (server-rendered, not locked behind JavaScript), and check that it is not buried many clicks deep. If a crawler cannot read it, nothing else matters.
2. Lead with a definition
Open the page — and ideally each major section — with a plain, self-contained statement of fact. A language model can lift a definition-first sentence verbatim with confidence. A page that opens with a marketing hook or a long preamble gives the model nothing clean to quote.
3. Add structured data
Mark up the page with JSON-LD schema that labels what it is: Article for editorial content, FAQPage for question-and-answer sections, HowTo for procedures. Schema removes ambiguity about the page's structure and makes specific passages safe to extract.
4. Use clear, question-shaped headings
Write headings as the questions a user would actually type or ask. This matches the page to the query and lets the model isolate the exact passage that answers a question instead of guessing. Keep the paragraph under each heading focused on that one point.
5. Cite your own sources
Pages that link to authoritative, verifiable sources read as more trustworthy — to readers and to models. Citing data, studies, and primary sources signals that your claims can be checked, which makes a model more comfortable repeating them.
6. Publish an llms.txt
Add an llms.txt file at your site root summarizing the site and pointing to your key pages. It gives AI systems a clean map of what you offer and which pages are authoritative on which topics.
7. Earn mentions elsewhere
Models weigh corroboration heavily. A claim that appears only on your own site is weaker than one echoed across reputable third-party sites. Coverage, citations, and mentions on other domains raise the odds that a model treats your page as a reliable source worth naming.
Then measure
Run the page through a GEO audit to confirm the structural signals — schema, definition-first opening, headings, llms.txt — are in place, and periodically ask ChatGPT your target questions to see whether your page is surfaced and cited.
Frequently asked questions
How does ChatGPT decide which pages to cite?+
When using its search or browsing mode, ChatGPT favors pages that are crawlable, clearly structured, state facts plainly, and are corroborated by other reputable sources. It cites passages it can extract and trust with confidence.
Do I need to allow AI crawlers?+
To be cited, yes. Check that robots.txt does not block crawlers like GPTBot and OAI-SearchBot — blocking them removes your pages from ChatGPT's citable index.
How long does it take to get cited?+
There is no fixed timeline. It depends on how often your page is recrawled, how authoritative your domain is, and how well-structured the content is. Structural fixes take effect on the next crawl; trust and corroboration build over months.
Does an llms.txt file guarantee citations?+
No. llms.txt helps AI systems understand your site, but citation still depends on content quality, structure, trust, and corroboration. It is one helpful signal, not a guarantee.
Can I check if my page is ChatGPT-ready?+
Yes. A GEO audit scores the structural signals that matter — crawlability, schema, definition-first structure, heading clarity, llms.txt — and returns a prioritized list of fixes.
See how your page scores
Run a free GEO audit and get a ranked list of fixes.