Find the real questions · Lesson 05

Build a query fan-out map

Turn one category into the decision questions AI must resolve, then choose which ones deserve pages or sections.

24 minWorked exampleField assignment
0%course
What you will leave with

A fan-out map with at least 12 questions and a content destination for each.

Answer engines rarely answer a buyer’s head term directly. They decompose it into hidden questions: suitability, proof, cost, risk, alternatives and constraints. Your content needs to resolve that bundle.

Worked example: law firm

personal injury lawyer
Injured workerCan I file if workers’ compensation already paid?What evidence should I save today?
ParentHow are child injury settlements approved?Who controls the settlement money?
Comparison shopperContingency fee vs hourly feeWhen should I not hire a lawyer?

Choose the right content destination

Add a sectionThe question supports the same intent and can be answered in 80–150 words.
Create a dedicated pageThe question has distinct intent, evidence, examples and a meaningful next action.
Create a data assetThe answer depends on facts competitors cannot simply paraphrase from you.
Do not publishThe question has no relevance, proof or business consequence.

Prompt you can reuse

You are mapping the questions an answer engine must resolve before recommending [CATEGORY].
Persona: [WHO]
Constraint: [BUDGET / LOCATION / PLATFORM / RISK]
Generate 5 problem, 5 comparison, 5 proof and 5 objection questions.
Do not write keywords. Write complete buyer questions.
For each question, recommend: existing section, new page, data asset, or do not publish.

Bad fan-out versus useful fan-out

Keyword variationsbest CRM / top CRM / CRM software

Same intent wearing three labels.

Decision coverageWhich CRM preserves attribution when a lead changes devices?

A constraint that requires evidence and changes the choice.

Do the work

Field assignment

Scenario

A law firm tracks only “personal injury lawyer.” Build the questions a worried prospect actually asks.

  1. List three personas and their immediate constraints.

  2. Generate problem, comparison, proof and risk questions for each.

  3. Cluster overlapping questions into one page, separate section or separate asset.

Definition of done

A fan-out map with at least 12 questions and a content destination for each.