What are fan-out queries?

Fan-out queries are the sub-questions an AI model generates internally while researching its answer to a user prompt. They represent the decomposition of a broad question into narrower information needs. If your content directly answers these sub-questions, you are more likely to be cited as a source.

How much does a request cost?

Each request costs approximately $0.10 via the DataForSEO LLM Mentions Search endpoint, regardless of whether you search by keyword or by domain.

What does the frequency ranking mean?

Fan-out queries are ranked by how often they appear across different parent questions. A higher frequency means the sub-question surfaces under multiple parent prompts, making it a higher-priority content target because answering it increases your citation chances across several queries.

What is the results limit slider for?

The slider (range 10 to 500, default 100) controls how many fan-out query results the API returns. A higher limit gives more comprehensive coverage but still counts as a single API call at the same cost.

Can I search by domain instead of keyword?

Yes. The tool supports both Keyword and Domain search modes. Domain mode returns the fan-out queries that led AI models to cite pages on that domain, which is useful for understanding why a competitor is being referenced.

How should I use the parent question mapping?

Each fan-out query maps back to the parent question that generated it. Use this mapping to build content clusters: group fan-out queries by their parent, then ensure your content answers the full set of sub-questions beneath each parent prompt.

Back to Tools

Fan-Out Query Explorer

Use cases

Planning content that answers the sub-questions AI uses to build answers Discovering content gaps AI systems expose in your topic coverage Understanding how AI breaks complex queries into research steps Prioritising content creation by fan-out frequency

Surfaces the fan-out queries (sub-questions) that ChatGPT generates when researching answers to queries in your topic space.

Queries the DataForSEO LLM Mentions Search endpoint by keyword or domain, deduplicates discovered fan-out queries, and ranks them by frequency.

Optionally clusters results by text similarity when scikit-learn is available.

Fan-out queries are only returned by the ChatGPT platform; Google AI Overviews do not provide them.

Platform

Python script (requires Python 3.x)

Input

DataForSEO API credentials (login and password)

A keyword or domain to explore

Location, language, and results limit settings

Output

Fan-out queries ranked by frequency with parent question mapping, plus a parent questions table with fan-out counts, AI search volume, and source counts.

View Source

Features

Search by keyword topic or by domain mentions
ChatGPT platform (the only source that returns fan-out queries)
12 location and 8 language options
Fan-out query deduplication and frequency ranking
Optional TF-IDF clustering when scikit-learn is available (agglomerative, threshold 0.7)
Parent question mapping (which questions spawned each fan-out)
Summary metrics: parent questions, total fan-outs, unique fan-outs, average per question
CSV downloads for fan-out queries and parent questions

How to use

1 Enter DataForSEO credentials in the sidebar
2 Select target type (Keyword or Domain)
3 Choose location and language
4 Enter your keyword or domain
5 Click Explore Fan-Out Queries
6 Review fan-out queries ranked by frequency
7 Download fan-out queries or parent questions CSV

Frequently asked questions

What are fan-out queries?: Fan-out queries are the sub-questions an AI model generates internally while researching its answer to a user prompt. They represent the decomposition of a broad question into narrower information needs. If your content directly answers these sub-questions, you are more likely to be cited as a source.
How much does a request cost?: Each request costs approximately $0.10 via the DataForSEO LLM Mentions Search endpoint, regardless of whether you search by keyword or by domain.
What does the frequency ranking mean?: Fan-out queries are ranked by how often they appear across different parent questions. A higher frequency means the sub-question surfaces under multiple parent prompts, making it a higher-priority content target because answering it increases your citation chances across several queries.
What is the results limit slider for?: The slider (range 10 to 500, default 100) controls how many fan-out query results the API returns. A higher limit gives more comprehensive coverage but still counts as a single API call at the same cost.
Can I search by domain instead of keyword?: Yes. The tool supports both Keyword and Domain search modes. Domain mode returns the fan-out queries that led AI models to cite pages on that domain, which is useful for understanding why a competitor is being referenced.
How should I use the parent question mapping?: Each fan-out query maps back to the parent question that generated it. Use this mapping to build content clusters: group fan-out queries by their parent, then ensure your content answers the full set of sub-questions beneath each parent prompt.

Want me to run this for you?

I run this tool as a managed service, or build something custom around your data. You get the insights without touching the code.

Book a Call

AI vs Classic VolumeScriptAI Search

Compare AI search volume against traditional Google volume per keyword.

AI Visibility CheckerScriptAI Search

See who gets cited in Google AI Overviews and ChatGPT for your topics.

BCG Matrix GeneratorScriptReporting

Generate BCG matrix visualisations from GA landing page reports with cascading folder analysis.

BERTlinkerSaaSInternal Linking

SaaS platform for smarter internal linking using ML embeddings and semantic analysis.

Best Sellers to XML SitemapScriptE-commerce

Generate XML sitemaps from GA landing page exports filtered by top transactions or revenue.

Hreflang Tag GeneratorBrowserTechnical SEO

Generate correct hreflang tags for international SEO.

Need something built for your business?

This tool started as bespoke client work. I build custom scripts, data pipelines, and full apps for SEO and product data problems that off-the-shelf tools don't solve.