Query fan-out estimator
A query fan-out estimator guesses which web searches an AI platform will run for a question, so you can write content that covers them.
A query fan-out estimator is a tool that guesses which web searches an AI platform will run when someone asks a question. You give it a topic or a question, and it lists the searches the platform is likely to make before it answers.
It exists because of query fan-out. Platforms like ChatGPT and Google AI Mode often turn one question into several searches of their own. Pages that rank for those searches can shape the answer, even if the person never typed them. An estimator helps you plan content for searches you can't see in advance.
How it's used
You'd usually run one before writing or updating a page. For example, Kestrel Roasters plans a page on "best coffee for a French press." An estimator might suggest the platform will also search for "French press grind size," "medium vs dark roast French press" and "French press coffee to water ratio." Kestrel adds a short section on each, so the page answers what the platform is likely to look up.
What to keep in mind
An estimate is a guess, usually made by a language model reasoning about what it would search. Real platforms fan out differently from day to day, and differently from each other. So treat the list as ideas to cover, and check it against the searches platforms actually run when you can see them.
Don't stuff a page with every search on the list either. Cover the ones a reader would want answered on that page, and link to other pages for the rest.
In AnswerOrbit
We don't estimate searches. We keep the ones the platforms actually ran, when they show them: ChatGPT, Perplexity and Copilot do, and Gemini, AI Overviews and AI Mode don't show theirs to us yet. Open an answer in Answers to see its searches. The Searches tab in Sources shows how often each platform searched and how often you were named when it did.