Query fan-out
Query fan-out is when an AI platform turns one question into several web searches of its own before it writes the answer.
Query fan-out is when an AI platform takes one question and runs several web searches of its own before answering. The person asks one thing. Behind the scenes, the platform searches for five or ten related things, reads the results, and writes one answer from all of them.
Google uses the term for how AI Mode and AI Overviews work, in its guidance on AI features. Other platforms do something similar, even if they don't name it.
An example
Someone asks: "What's the best espresso machine for a small café?"
The platform might run separate searches for small café espresso machines, espresso machine reviews, two-group machine prices, and a couple of brands it already knows. Then it reads the top pages from each search and writes its answer.
So the answer can be shaped by pages that rank for searches the person never typed.
Why it matters
Fan-out widens the set of pages that can influence an answer. Ranking for the exact question isn't enough. You also want to show up for the searches the platform runs around it: comparisons, prices, reviews, and "best" lists in your category.
It also explains why answers vary. If the platform fans out differently on Tuesday than on Monday, it reads different pages, and the brands in the answer can change.
Query fan-out in AnswerOrbit
When a platform shows the searches it ran, we keep them. Open an answer in Answers to see its background searches. The Searches tab in Sources shows how often each platform searched before answering, and how often you were named when it did. We keep the searches from ChatGPT, Perplexity and Copilot. Gemini, AI Overviews and AI Mode don't show theirs to us yet.