What Type of Content Does ChatGPT Cite the Most?

This live report tracks the daily breakdown of the content types ChatGPT cites over the past 30 days: news articles, reviews, forums, documentation, and more. Use it to see which formats ChatGPT prefers and how that mix is shifting.

What Type of Content Gets Cited by OpenAI ChatGPT?

Based on Promptwatch data from Jan 1 - Feb 4, 2026. Covers 1,027,802 classified citations. The chart shows the percentage breakdown of citation content types per day. Only citations with a classified content type are included.

Citation Content Types Breakdown

• Listicle: 17.4%
• Landing Page: 16.7%
• Product Page: 14.4%
• News Article: 10.8%
• How-To: 5.9%
• Wiki: 4.4%
• Social Post: 3.6%
• Documentation: 2.3%
• FAQ: 1.6%
• Review: 1.5%
• Comparison: 1.5%
• Research Paper: 0.6%
• Press Release: 0.4%
• Opinion: 0.3%
• Case Study: 0.3%
• Glossary: 0.2%
• Video: 0.1%
• Q&A: 0.1%
• Forum Post: 0.1%
• Interview: 0.0%
• Podcast: 0.0%
• Infographic: 0.0%
• Comment: 0.0%

What this means for you

No single format owns ChatGPT's citations. Listicles (17% on average), landing pages (17%), and product pages (~14%) sit at the top, with news articles close behind at ~11%. That commercial pages rank this high is notable: ChatGPT frequently cites brand-owned pages directly rather than relying only on third-party editorial coverage.

The mix also moves. Over this five-week window, listicles climbed from roughly 13% to 21% of daily citations while landing pages slid from about 22% to 13%, so the two formats effectively swapped places. Whatever ChatGPT favors today is not guaranteed next month.

How to act on it

• Turn your best pages into list-style roundups and "best of" comparisons. Listicles are the single most cited format here, and their share is still growing.
• Don't neglect your commercial pages: landing and product pages together account for roughly 30% of ChatGPT's citations, so clear specs, pricing, and factual copy on those pages directly earn visibility.
• Re-check this mix monthly before committing a content calendar. A format that gained or lost 5+ points in a single month can flip your priorities.

How we collect this data

We collect millions of prompt responses, citations, and click data from the actual user interfaces of major AI platforms: over 26 billion data points and growing. This gives us one of the largest datasets on how AI search engines cite sources and recommend brands.

Real UI monitoring

Data straight from the interfaces of ChatGPT, Gemini, Perplexity, Claude, AI Overviews, and more.

26B+ data points

Over 26 billion analyzed citations, prompts, and responses, one of the largest AI search datasets available.

Continuously updated

Refreshed constantly so the trends you see reflect the latest behavior of AI search engines.

Aggregated & public

Published freely for the GEO community, based on aggregated, non-identifiable trends.