Features / Query fan-out analysis

Query fan-out analysis

One buyer question becomes several searches inside the engine. Fan-out analysis shows you those hidden sub-queries and the sources pulled in to answer each one.

Free, no card, first read in minutes

The problem

You optimise for the question you think your buyer asked. The engine asked something else first.

Behind a single prompt, a model decomposes the request into narrower searches, answers each, then assembles a response. If none of your pages match those narrower searches, you are absent from the answer no matter how well your main page is written.

How it works

We capture the decomposition For each tracked prompt, we record the narrower searches the engine ran on its way to an answer.

Sources are attached per sub-query Each sub-query carries the URLs cited to answer it, so you can see who owns that specific question.

Gaps are surfaced Sub-queries where none of your pages appear are the clearest content brief you will get.

Compare across engines The same prompt fans out differently per engine. Where they overlap is where a single page does the most work.

What you see

01 The sub-queries generated behind each prompt you track

02 Which sources were cited to answer each one

03 The sub-queries where you have no matching page at all

04 How fan-out differs between engines for the same prompt

05 Which of your pages already answer a sub-query well

Who it's for

Content leads who need next quarter’s brief to come from what engines actually search, rather than from a keyword tool.

Fan-outone prompt

Prompt "Best AEO platform for a B2B SaaS brand"

└

"AEO platform pricing comparison"4 sources cited · none yours

└

"AEO tools for agencies reddit"6 sources cited · none yours

└

"what is answer engine optimization"3 sources cited · 1 yours

└

"best AI visibility tracking software 2026"7 sources cited · none yours

4 sub-queries, 3 with no page of yours

What fan-out changes about content

It moves the unit of work from the topic to the question.

Brief Write for the sub-query A page that directly answers one narrower search is cited more often than a broad page that covers everything loosely.

Priority Chase the overlap Sub-queries that appear behind several of your prompts, and across several engines, are worth a page of their own.

Proof Some gaps are not yours to fill Where a sub-query is answered entirely by community sources, a landing page will not close it. That is a Reddit job.

Frequently asked questions

Is fan-out the same as related searches? No. Related searches are suggestions shown to a user. Fan-out is what the model itself searched internally before writing its answer, which is why the sources attached to it matter so much.

Does every engine fan out? The behaviour and depth vary by engine, which is part of what the view shows. Google AI Mode and Perplexity tend to decompose more aggressively.

How do I use this? Take the sub-queries where you have no cited page, check whether the cited sources are owned, community or third-party, and brief the work into that channel.

Can I export it? Yes, as CSV, per prompt.

Feature AI citation tracking Every source an engine cited, sorted by channel. Read more › Feature Citation Impact The handful of pages carrying your presence in AI answers. Read more ›

See what the engine actually searched

See the sub-queries running behind your category.