Platform guide
How to get cited by ChatGPT
To get cited by ChatGPT, publish clear, factual pages that answer a specific question directly, use FAQ schema and comparison tables, build topical authority through internal links, and keep content fresh with recent dateModified values. ChatGPT cites sources it trusts: clear, factual, well-structured pages that directly answer the user's question.
1. Answer questions directly
Put the answer near the top. ChatGPT extracts concise statements. If your page buries the answer under paragraphs of context, it is less likely to be cited.
2. Use structured formats
- ✓ FAQ schema
- ✓ Comparison tables
- ✓ Numbered steps and checklists
- ✓ Clear H2 and H3 headings
3. Build topical authority
Cover your category thoroughly. A single page is rarely enough. Create a cluster of related pages that link to each other: definitions, comparisons, how-tos, and benchmarks.
4. Keep content fresh
ChatGPT's knowledge cutoff and browse behavior favor recent, authoritative sources. Update key pages regularly and add dateModified schema.
Frequently asked questions
How do I get my website cited by ChatGPT?
Allow the OpenAI crawlers in robots.txt, answer the page's own question in one sentence near the top, structure the rest into extractable chunks with question-form headings and FAQ schema, and earn mentions on third-party pages that ChatGPT already trusts. Retrieval first — a blocked crawler makes everything else irrelevant.
Which crawlers does ChatGPT use?
Three separate user agents, and they do different jobs: GPTBot collects content that may inform model training, OAI-SearchBot powers ChatGPT's search index, and ChatGPT-User fetches a page live when a user's question triggers a browse. Blocking any one of them costs you a distinct surface, so decide on each deliberately rather than disallowing them as a group.
Why does ChatGPT cite my competitor instead of me?
Usually one of three reasons. It cannot reach your page — crawler blocked, or the content is rendered client-side only. It reached the page but found nothing self-contained to quote. Or the answer is coming from training data rather than a live fetch, in which case it reflects what the open web said about your category months ago, and third-party mentions move it, not your own site.
Does ChatGPT use my llms.txt file?
OpenAI has not committed to reading llms.txt, so treat it as optional rather than a lever. It costs almost nothing to publish and it does help agents and some crawlers find your primary pages, but robots.txt access and server-rendered content are the checks that actually determine whether ChatGPT can quote you.
How long does it take ChatGPT to pick up a new page?
For browse-enabled answers, days after the page is crawled. For answers drawn from training data, the next model release — months, and outside your control. This is why the same change can appear to work in one conversation and do nothing in another: they are two different retrieval paths.
Can I track whether ChatGPT mentions my brand?
Yes, but not through analytics — a generated answer produces no impression and often no click. You track it by running your buyers' prompts against ChatGPT on a schedule and recording whether your brand appears, where, with what sentiment, and which URL was cited.