How do I write content that ChatGPT will cite?
ChatGPT cites content that answers a specific question in 1-3 sentences and comes from a domain it trusts. Five tactics: open every section with the answer (BLUF), use question-shaped H2s, ship FAQPage schema, structure data as HTML tables, and republish with fresh dateModified when claims change.
ChatGPT does not cite long prose narratives. It extracts answer-shaped passages that fit cleanly into its response. Five tactics that move the needle.
Tactic 1: BLUF (Bottom Line Up Front)
Every H2 section opens with the answer in 1-3 sentences. Elaboration follows. This mirrors how ChatGPT extracts: the first paragraph of each section is the candidate citation. Bury the answer in paragraph 3 and you lose the snippet.
Tactic 2: Question-shaped H2s
Write headings as questions readers might ask AI. "What is the cost of vendor lock-in?" beats "Cost considerations." Question-shaped H2s match the actual prompt shape ChatGPT users type, increasing extraction probability.
Tactic 3: FAQPage schema
A FAQPage JSON-LD block with 8+ Q&A pairs gives ChatGPT (and Claude, Perplexity, Gemini) a structured map of what your page answers. Each Q+A becomes an extractable unit. Pages with FAQPage schema get cited 2-3x more often than pages without, per published GEO research.
Tactic 4: HTML tables for comparisons
ChatGPT extracts HTML tables nearly verbatim. If you write a comparison as prose ("X is better than Y because..."), ChatGPT rewords. If you write a table with X, Y, and 5 dimensions of comparison, ChatGPT extracts the table structure intact.
Rule of thumb: if your answer has 3+ items that vary across 2+ dimensions, it is a table. Markdown tables convert to HTML in most stacks (MDX, Contentlayer, standard markdown).
Tactic 5: Republish with fresh dateModified
ChatGPT browsing favors fresh sources for evergreen queries. When pricing, product features, or industry data changes, update the relevant content AND update Article.dateModified in the schema. ChatGPT re-prioritizes the updated URL on next crawl.
How to test which of your pages ChatGPT cites
Radar Answer Engine Citation Tester queries ChatGPT, Claude, Perplexity, and Gemini against the prompts you specify, then reports which URLs each engine cited. You see in 60 seconds whether your page made the citation list — and which competitor pages beat you.
Want to take action on this?
Related questions
What is Answer Engine Optimization (AEO)?
Answer Engine Optimization (AEO) is the practice of structuring web content so AI engines can extract and cite it as a complete answer. AEO focuses on extraction-friendly formatting: BLUF paragraphs, FAQPage schema, HTML comparison tables, and speakable schema.
How-toHow do I get cited by Perplexity?
Perplexity cites sources it crawls + ranks for the live query. To get cited: allow PerplexityBot in robots.txt, ship answer-first content (FAQPage schema, BLUF paragraphs), and acquire citations from high-authority domains Perplexity already ranks.
DefinitionWhat is Generative Engine Optimization (GEO)?
Generative Engine Optimization (GEO) is the practice of structuring web content so AI search engines like ChatGPT, Claude, Perplexity, and Gemini cite it in their responses. Unlike SEO which optimizes for keyword rankings, GEO optimizes for entity recognition, structured data, and citation probability.
WhyWhy does Perplexity cite my competitor instead of me?
Perplexity ranks sources per-query by domain authority, content relevance, and recency. If your competitor is cited and you are not, they typically beat you on one of three signals: answer-first content structure, citation accumulation from high-trust domains, or recency of updates on the relevant page.
