🎯 Why this view exists
To build a response, models often generate multiple internal queries, known as fan-out queries, to explore the topic before producing the final answer.
These fan-out queries largely determine:
which angles of the topic are explored,
which domains and URLs are retrieved,
and whether your domain or brand appears later as a Source or a Link.
The Fan-out Queries view is designed to make this internal exploration visible, so you can better understand how generative answers are constructed.
To go further:
📊 What this view helps you analyse
The Fan-out Queries view helps you analyse:
how LLMs decompose a prompt into multiple internal queries,
how stable or volatile these fan-out queries are,
how this behaviour differs across engines,
and how your domain and competitors are positioned within this fan-out layer.
It acts as a bridge between LLM reasoning and search visibility signals.
#1 Overview tab: identifying global fan-out patterns
#1 Overview tab: identifying global fan-out patterns
The Overview provides a high-level dashboard to quickly identify patterns and trends across fan-out queries.
Here, you can observe:
Coverage: the percentage of fan-out queries for which your domain is present when URLs are retrieved
Total QFO: the total number of fan-out queries generated across tracked prompts
Average QFO: the average number of fan-out queries per prompt, by engine
You can also analyse coverage distribution:
by thematic or topic
by intent type (exploratory, comparative, decision-making, etc.)
This view is designed to answer questions like:
How broad is the fan-out explored by the model?
On which topics or intent types is my coverage stronger or weaker?
Do different engines behave differently at a high level?
#2 Fan-out tab: analysing individual fan-out queries
#2 Fan-out tab: analysing individual fan-out queries
This view provides the detailed list of fan-out queries, either:
shown flat ("Grouped by: default")
or grouped by prompt, to compare fan-out patterns across engines.
For each fan-out query, you can see:
the fan-out query itself
the intent(s) associated with it
its stability over the last 12 collects for a given prompt and engine: 1 bar is the fan-out was present in the collect, no bar if it was absent.
whether your domain is present in the listed URLs provided by the model
the full list of URLs, ordered as returned by the API
This makes it possible to:
understand which internal queries the model repeatedly relies on,
distinguish structural fan-out queries from more occasional ones,
and observe relative positioning versus competitors within a given fan-out.
Reminder: Depending on the model and API capabilities, fan-out visibility may be partial and reflects the retrievable part of the model’s exploration, not its full internal reasoning.
#3 Fan-out at prompt level: understanding how a single prompt is decomposed
#3 Fan-out at prompt level: understanding how a single prompt is decomposed
From the Rankings or Search Queries views, you can access the above fan-out details for a specific prompt.
This perspective helps you understand:
how a single prompt is broken down into multiple internal questions,
which types of fan-out queries the model prioritises for that prompt,
and how your presence varies across these internal queries.
It provides a focused lens on prompt-level reasoning, complementary to the global overview.
💡 Reading cues: how to interpret fan-out data
These patterns are signals to read, not rules.
Fan-out queries that are stable over time often reflect core questions the model consistently asks itself.
Differences across engines highlight engine-specific reasoning strategies.
Fan-out queries where competitors are consistently present help explain why certain domains repeatedly influence answers.
Recurring fan-out queries can also be read as indicators of topics the model repeatedly expects to find content about when building answers.
Fan-out data should be read as a map of the model’s internal exploration, not as a traditional ranking report. It highlights patterns of exploration, not direct causality between a single query and final visibility.
✨ How to use this view effectively
Start by identifying fan-out queries that appear consistently across engines and collects. These recurring queries often reflect the core questions the model relies on to explore a topic and determine which domains influence the final answer.