How to Track All Fan-Out Queries at Once in AmICited

The fan-out queries AmICited discovers for a prompt are real questions AI engines research, so they’re worth tracking in their own right. The Track all prompts button turns the whole fan-out set into monitored prompts in one click.

Quick Steps

• Open a tracked prompt’s detail page and find the Fan-out queries panel.
• Click Track all prompts (N) to add every discovered sub-query as its own tracked prompt, or use + Track to add one at a time.
• Tracked rows show a ✓ Tracked state so you can see coverage at a glance.
• Each tracked sub-query gets its own run history, citation results, and share of voice contribution.
• Set schedule and engines for the new prompts afterward in the Prompts table, and mind your run quota when tracking large sets.

What is query fan-out?

When you type a question into ChatGPT, Perplexity, Gemini, or a Google AI Overview, the engine rarely answers from that single sentence alone. Instead, it silently rewrites your prompt into a batch of related sub-queries, runs research against each one, and stitches the results together into the answer you see. That decomposition process is called query fanout : a single user prompt “fans out” into anywhere from a handful to dozens of narrower questions, each one probing a different angle: a definition, a comparison, a price point, a use case, a competitor name.

This matters for AI search visibility because citations don’t just happen at the level of the original prompt: they happen at the level of every sub-query the engine generates along the way. A brand that’s invisible in the answer to “best project management software” might still get cited in one of the fan-out sub-queries feeding that answer, like “project management software for remote teams” or “project management software pricing comparison.” Conversely, ranking well on the parent prompt doesn’t guarantee coverage across the sub-queries an engine explores underneath it. If you only track the prompts you originally thought to write, you’re missing the actual research trail the AI model followed to construct its response.

This is also why generative engine optimization treats prompt sets differently from traditional keyword lists. A keyword list is static and chosen by a marketer; a fan-out set is dynamic and chosen by the AI engine itself, which means it reflects how the model actually reasons about a topic right now. Tracking those sub-queries gives you a more honest picture of where your AI visibility gaps really are, because you’re measuring against the questions engines are actually asking, not just the ones you assumed they’d ask.

AmICited surfaces this fan-out set automatically for every prompt you track, then gives you a direct path to convert any or all of those sub-queries into prompts of your own, which is what the rest of this guide covers.

Where to find it

On the Fan-out queries panel of a prompt detail page. Open any tracked prompt and this panel lists the sub-queries AmICited identified as part of that prompt’s research trail. You have two options for turning them into tracked prompts:

• Track all prompts (N): the button in the panel header adds every fan-out query as a tracked prompt at once (the number shows how many).
• + Track: the button at the end of each row adds just that single sub-query.

Once added, rows show a ✓ Tracked state so you can see at a glance what’s already being monitored and what’s still just a discovered-but-untracked query. This makes the panel double as a running audit of your prompt coverage for that topic: everything without a checkmark is a question engines are researching that you’re not yet watching.

What it measures

Each fan-out query, once tracked, becomes a full prompt in your AmICited account, meaning it gets its own run history, its own citation results across engines, and its own contribution to your overall share of voice . Rather than inferring your standing on a sub-query indirectly through the parent prompt’s result, you get a direct, engine-by-engine read on whether your brand is cited for that specific angle.

This is where fan-out tracking pays off over time. A single parent prompt might expand into sub-queries about pricing, integrations, alternatives, and use-case fit. If your brand shows up reliably on the “alternatives” and “integrations” angles but drops out of the “pricing” and “use-case fit” ones, that’s a concrete, actionable pattern: you know exactly which type of content or page is under-serving AI engines, rather than just knowing your visibility on the topic overall is mediocre. Tracked at scale across many prompts, fan-out sub-queries function as an AI rank tracker for the long tail of questions beneath your core topics, the layer of specificity that’s usually invisible if you only monitor the prompts you thought of yourself.

How to use it

1. Review the list first. Skim the fan-out queries and their type tags: most will be relevant, but you can add them selectively with the per-row + Track if you only want some. This is worth doing deliberately rather than reflexively tracking everything: a sub-query that’s only tangentially related to your business will dilute your prompt set with noise rather than signal.
2. Track all for full coverage. If the prompt is important (a core product category, a competitive term, a high-intent buying question) adding the whole set gives you visibility into every angle engines explore around it, not just the one you happened to write. This is the fastest way to go from a single hand-picked prompt to a genuinely comprehensive AI prompt tracking set for that topic.
3. Set their schedule and engines afterward in the Prompts table. The new prompts inherit sensible defaults, but you can adjust them together using bulk actions , for example, tightening the run frequency on a cluster of high-value sub-queries or aligning which engines they check against the rest of your set.
4. Mind your quota. Adding many prompts increases how many runs you use, so track the fan-out sets for your highest-value prompts first, and treat the per-row + Track option as your filter for lower-priority topics where full coverage isn’t worth the run budget.

Why this closes a real gap in AI visibility monitoring

Most teams that start monitoring AI search visibility begin with a prompt list built by guesswork: a handful of questions someone on the team assumes customers or AI engines would ask. That list is a reasonable starting point, but it’s inherently limited by what a human thought to write down. Query fan-out tracking replaces guesswork with observed behavior: instead of imagining what an engine might research, you’re looking directly at what it did research the last time it answered a prompt you care about.

Doing this consistently also changes how you read your citation data. A single tracked prompt gives you one data point per engine per run. A fully tracked fan-out cluster gives you a small dataset (sometimes a dozen or more related queries) that you can read as a pattern. That pattern is far more useful for diagnosing why AI Overviews or ChatGPT cite a competitor instead of you: it tells you whether the gap is topical (you’re missing content on one specific angle) or structural (your content exists but isn’t being selected as a source across the board).

Where to go next

Once you’ve tracked a fan-out set, treat it the same way you’d treat any other prompt cluster in AmICited: watch how citation tracking trends across the sub-queries over the following weeks, and use the pattern of wins and gaps to prioritize what content to fix or publish next. If you’re building out this workflow prompt by prompt across a large site or a multi-brand portfolio, it’s worth pairing fan-out tracking with a broader look at which prompts trigger AI to mention your brand in the first place, so your core prompt list and its fan-out expansions stay aligned with what actually drives citations rather than what you assumed would.