Concepts
Query fan-out
Query fan-out is what happens when an assistant takes one question and expands it into several internal searches before composing an answer. The buyer asks one thing; the engine runs many. Which pages get retrieved, and which of those get cited, is settled across those hidden sub-queries rather than the visible one.
Why it complicates measurement
The prompt you tracked is not the query that fetched the sources. A page can be retrieved for a sub-question you never thought to measure, and the relationship between the wording of a prompt and the wording that actually did the retrieving is not something you can observe from outside.
What it changes in practice
Content that answers one narrow question completely tends to do better than content covering a topic broadly, because it matches a sub-query cleanly. It is also an argument for measuring a wide set of prompts rather than repeating a narrow one very often.
One question, several searches
A buyer asks which tool suits a small team on a tight budget. Before answering, the engine may search separately for tools in the category, for pricing, and for small-team suitability, then compose one answer from what each returned. A page can be retrieved for the pricing sub-question and never be seen by the prompt you actually tracked.
Related terms
- Answer engine — The systems that do the expanding.
- Citation — Sources are chosen across the hidden sub-queries, not the visible one.
- Prompt taxonomy — An argument for tracking many prompts rather than repeating a few.
Last reviewed . Definitions are reviewed quarterly, and whenever the underlying measurement changes.
Don't let AI decide your brand's future without you.
See exactly where you stand vs. competitors—and what to do about it.