How should I use the content type labels?

As a gap check. If a sub-query points at a comparison and you have no comparison page, that sub-query is being answered by someone else. The labels tell you which format each part of the fan rewards: docs, guides, reviews, or forum presence.

Why do some fan-outs skip the comparison group?

Not every intent has a plausible comparison angle. A purely informational or local query fans out into reformulations and follow-ups instead; the simulator only includes groups that fit the intent.

Can I share a result?

Yes. The share button copies a link that prefills your query for whoever opens it; they run the simulation themselves with one click.

What does AI search really ask?

AI search engines never run your query alone: they expand it into a fan of sub-queries and answer from all of them. Type a buyer question and see a realistic fan-out, with the content type each sub-query rewards.

how it works

One query in, a research plan out

Enter a query

A buyer question works best: the kind of thing someone types into Google AI Mode or ChatGPT when researching a purchase.

See the fan-out

The simulator produces 8 to 14 sub-queries grouped by intent: reformulations, implicit questions, comparisons, follow-ups, and related entities.

Match content to sub-queries

Each sub-query is labeled with the content type most likely to answer it: comparison pages, docs, reviews, forums. That is your content plan for the topic.

What is query fan-out?

When you ask an AI search engine a question, it quietly issues several related searches, reads results for each, and composes one answer. Google describes this technique in AI Mode; ChatGPT search and Perplexity behave the same way. The consequence: your page competes on every sub-query in the fan, not only the one the user typed.

Why fan-out changes SEO math

Ranking for the head query is no longer the whole game. If the fan includes a comparison sub-query and you have no comparison page, a competitor gets cited even when the user never typed the word versus. Covering a topic now means covering its fan: the implicit questions, the alternatives angle, the follow-ups.

A simulation, not a leak

Engines do not publish the exact sub-queries they run, and the real fan varies by user, context, and day. This simulator models the documented behavior to show the shape and breadth of a realistic fan-out, which is what you need for content planning.

the real question

Which of these queries does your brand appear in?

Citlyze asks the engines your buyer prompts every day, records the answers, and shows which queries mention you, which cite you, and which belong to competitors.

Start the 14-day trial Book a demo

FAQ

Query fan-out, answered

Want a demo of Citlyze? Book a demo

It is a realistic model of the behavior, not a log of it. Engines do not expose their real sub-queries, and the fan varies per user and session. The simulator exists to show the shape of the expansion so you can plan content against it.

Does AI recommend you when buyers ask?

Track how ChatGPT, Gemini, Perplexity, DeepSeek, ByteDance, and Grok mention, cite, and rank your brand. Review tracked answers and the gaps to close.

Start free trial Book a demo

No credit card required