Free Tool

Query Fan-Out Previewer

Enter any topic or prompt. See the 8–16 sub-queries an AI search engine would fan it out into — grouped by intent, visualized as a live branching tree.

Heuristic preview — no live model call, no data sent anywhere

Free · runs in your browser · nothing sent to any server

Try an example
Search volume: 0

These sub-queries are machine-generated — they live inside the AI engine's retrieval layer, not in any keyword tool. Your pages get cited through queries nobody ever types. That's the point of covering the cluster.

12 sub-queries across 6 intent categories

Coverage check
Paste your page's text to see which sub-queries it already answers
0%
sub-queries covered → citations you're eligible for

Want a real model's decomposition instead of a heuristic preview?

Simulator prompt — paste into ChatGPT or Claude

Copy this and paste it into ChatGPT, Claude, or Gemini to get a real model's fan-out decomposition of your query.


        
Why this matters

Covering the cluster beats chasing one keyword

When someone asks an AI engine a question, the engine doesn't run that question as a search. It fans out into 8–16 parallel sub-queries, retrieves the best source for each, and synthesizes one answer with citations. Your page gets cited if it matches one of those sub-queries — not the prompt the human typed.

This breaks classic SEO intuition. The queries that actually retrieve your page have zero measurable search volume in any keyword tool — they're machine-generated on the fly. The implication: stop targeting one keyword. Map the full cluster of sub-questions an engine would generate, give each its own answer-first section, and you become eligible for citations across the whole fan-out.

This tool gives you a structural preview of that fan-out. It's heuristic — it applies known templates to your input — not a live model call. For a live model's actual decomposition, use the copyable simulator prompt above. For the full depth on the mechanic: read the companion deep-dive →

← Back to the AI search optimization field guide

Quick answers
What is query fan-out?
The retrieval technique where an AI search engine expands one user prompt into multiple related sub-queries — typically 8–16 — runs them in parallel against its index, and synthesizes a single answer from the combined results. Google uses the term for AI Mode; ChatGPT search, Perplexity, and Gemini use the same pattern.
How many sub-queries does AI search generate per prompt?
Standard AI Mode-style fan-out generates 8–16 sub-queries per prompt. Deep-research modes — Gemini Deep Research, ChatGPT deep research, Claude research — generate dozens, sometimes 50 or more, across multiple iterative steps in a single research run.
Is this a real model or a heuristic preview?
This is a heuristic previewer. It applies known fan-out templates to your input — definitional, comparative, procedural, transactional, evaluative, and edge-case — to give you a structural preview of how an engine would likely decompose your query. It's not a live model call. For a real model's decomposition, use the copyable simulator prompt in the tool to run against ChatGPT or Claude.
Does this send my data anywhere?
No. The entire tool runs in your browser — no API calls, no servers, no network requests for query processing. Your input never leaves your device. The only external requests are Google Fonts on page load and the standard GA4 analytics snippet.
Put it to work

Map the cluster. Earn the citations.

The fan-out preview shows you which sub-questions to cover. LumenGEO shows you whether ChatGPT and Perplexity are actually citing you right now — with live AI queries and competitor tracking.

Run a full AI-citation audit on LumenGEO → Work with me