AI search engines demonstrate significant divergence in content citation and preference, with ChatGPT and Claude sharing only approximately 8% of their cited sources for identical queries, according to Profound data.
ChatGPT heavily indexes community content such as Reddit and Quora, which constitute about 16% of its citations, while Claude utilizes this type of content less than 1%. Conversely, Claude shows a preference for listicles, making up 36% of its citations compared to ChatGPT's approximately 20%, and favors opinion content at 13.2% versus ChatGPT's 7.2%.
This disparity extends to their relationship with traditional search engines; 64% of the websites Claude cites also appear in Google's top 50 results for the same query, whereas only 37% of ChatGPT's cited sites do. This suggests that traditional SEO efforts may improve visibility for Claude but are less effective for ChatGPT.
Consequently, marketers must recognize that "AI visibility" is not uniform across different large language models (LLMs) and requires tailored strategies based on target audience engagement. Tracking visibility by individual AI engine is crucial, as efforts that yield results in one platform may be ineffective in another.
User-generated content and community engagement are key drivers for ChatGPT, while listicles and traditional search rankings are more impactful for Claude.
Sources and references
Original report
7 AI search shifts you can’t afford to ignore
Search Engine Land
Read on Search Engine Land (opens in a new tab)Ref 003PBBSS
Follow this story
Get Artificial Intelligence in your inbox — new stories in 40 words or fewer, once a week.
Subscribe