How to use the Fan-Out Query Analyzer
See the hidden background searches AI models fire when evaluating your category — and build content that captures them.
When an AI model receives a query like "what's the best CRM for small businesses?", it doesn't answer from memory alone. It fires a set of background sub-queries to gather specific, current information before composing a response. These are called fan-out queries, and they're the searches that determine which businesses actually appear in AI recommendations. The Fan-Out Query Analyzer makes them visible.
Running the analyzer
Enter any query your potential customer might ask an AI model, a category, a problem, or a comparison. The analyzer processes it across four models: ChatGPT, Claude, Perplexity, and Gemini. Results appear within a few seconds and show every background query each model generates, along with the cross-model consensus score for each one.
Reading the results
Each fan-out query displays a consensus percentage showing how many of the four AI models fire that same sub-query. A query with 100% consensus means all four models are searching for that specific information when answering your input, making it a critical content gap if your site doesn't address it. Lower consensus queries (25–50%) are still worth covering but are lower priority.
The model columns show which specific models fire each query, so you can see where ChatGPT and Claude diverge from Perplexity and Gemini. If your primary concern is ChatGPT recommendations, focus first on queries that ChatGPT fires consistently.
What to do with the results
Each fan-out query is a content gap. If your site doesn't answer that sub-query directly, AI models will find a competitor that does and cite them instead. For high-consensus queries, create dedicated pages or clearly structured FAQ sections that answer the question explicitly. For comparison queries (e.g. "HubSpot vs Salesforce pricing"), structured comparison content on your site gives models a direct source to cite.
After publishing new content, re-run the analyzer to verify the queries haven't shifted, and then run an AI Readiness Scan to confirm your new content is discoverable.
How this differs from keyword research
Standard keyword research shows what humans type into Google. Fan-out query analysis shows what AI models search for when constructing their own responses. These are often completely different. Fan-out queries are typically more specific, structured around comparison, validation, and evidence-gathering rather than discovery. A site can rank well on Google and still be nearly invisible to AI models if it doesn't answer the sub-queries they rely on.