1.2M citations analyzed across ChatGPT, Claude, Gemini, and Perplexity. 68% of all citations came from just 400 source domains. The ranked list is not what you'd expect, and what it means for your content strategy is specific and actionable.

The 400 domains

Seven domain types dominate: UGC platforms (Reddit, Quora, Stack Overflow), review aggregators (G2, Capterra, Trustpilot), first-party vendor pages structured for extractability, category-specific newsletters and blogs, general-interest news publishers, Wikipedia and Wikidata, and academic or institutional sources. The mix varies sharply by engine and by category, Perplexity cites Reddit at nearly twice the rate of ChatGPT in B2B software categories.

What it means for your content plan

If your brand is not on the 400, if none of your content surfaces in the citation stack for your category's key prompts, you're in the 60%: the long tail that gets cited occasionally, doesn't compound, and has no systematic way to influence what the engines say about you. The paper walks through how to identify which of the 400 matter for your specific category and the exact sequence of actions to earn presence in them.

// try it yourself
See what the engines say about you.

Start a free Promptive trial, 7 days of full data, no card required, five minutes to first insight.