One query becomes twenty.
See them all.
Free query fan-out checker. See the real sub-queries Gemini runs behind every AI answer.
This reads the real sub-queries Gemini runs when it answers a prompt. Observed in the grounding data, not predicted by the model. Type a query and see how AI breaks it apart.
We run your query through Gemini with live Google Search grounding and show you what it actually searched.
Ranking first for the head term stops being enough.
AI does not answer your keyword. It breaks the question into the sub-queries above and answers each one, then cites whoever covered it best. You can own the head term and still go unmentioned if your site never answers the parts that fan out from it.
Every sub-query is a page you either answer or hand to a competitor.
Questions about the tool.
What is query fan-out in AI search?
Query fan-out is the process where an AI engine like Gemini takes one user prompt and silently runs multiple background searches against it, then combines the answers from each one to produce a single response. It's how AI Mode and Google AI Overviews handle complex questions: not by searching for your exact phrase, but by breaking it into smaller sub-queries and answering each one separately. The pages that get cited are the ones that answer the sub-queries cleanly, not necessarily the ones ranking for the head term.
Are these the actual sub-queries Gemini ran, or guesses?
Actual. The tool reads webSearchQueries straight from the grounding metadata Gemini returns. We don't ask the model to tell us what it would search, we read what it did search.
Which model does it use?
Gemini 2.5 Flash with Google's Search grounding tool enabled, which is the same Search tool Google uses internally for AI Overviews.
Why is the sub-query list sometimes empty?
If Gemini answered from its own training data without running a grounded search, there are no sub-queries to surface. Try a more specific or current-events query and the fan-out usually appears.
Does this also work for ChatGPT or Perplexity?
Not directly. Only Gemini exposes the real sub-queries through grounding metadata. ChatGPT and Perplexity do not surface this publicly. The pattern Gemini fans out to is still a useful proxy for how AI search breaks down a topic in general.
Is my search query stored anywhere?
Nothing is logged on our side. Queries do hit Google's Gemini API, so Google's API terms apply on their end.
How is this different from Google's 'People also ask'?
'People also ask' reflects how people search on Google. Fan-out queries are how a generative AI breaks your query down internally before it searches the web. Different layer, useful for different things.
Want to know which of these sub-queries your site actually answers?
That is the audit. I map your pages against the fan-out and show you exactly where the gaps are, and which ones are worth closing first.