# AI retrieval fan-out mapper Human Guide

## What This Is For
Maps the search queries and concrete fact needs that answer engines may generate behind a user prompt, then turns them into evidence-backed content requirements. It gives the agent a clearer input/output frame for AI retrieval fan-out mapper: what context to ask for, what decisions to make, and what usable artifact to return.

Use this as a human-readable version of the AI retrieval fan-out mapper agent skill. It is meant for marketers, operators, founders, and other non-coders who want the workflow without reading agent-specific implementation instructions.

## When To Use This
- Use this when you need a repeatable process for AI retrieval fan-out mapper.
- Use this when the task needs judgment, examples, constraints, or a clear output format rather than a one-off prompt.
- Use this when you want to hand an AI assistant enough context to produce a usable marketing artifact.

## When Not To Use This
- Do not use this when you only need a quick factual answer.
- Do not use this when the work depends on private data you cannot share with the assistant.
- Do not use this as a replacement for legal, compliance, financial, or medical review.

## What You Need Before Starting
- The goal or business outcome you want.
- The audience, customer segment, or market context.
- Any source material the assistant should respect, such as notes, briefs, examples, URLs, or brand guidance.
- Constraints such as tone, length, channel, deadline, region, or approval requirements.
- A clear definition of what a good final answer should look like.

## Step-By-Step Workflow
1. State the job clearly: "Use the AI retrieval fan-out mapper guide to help me with..."
2. Add context: audience, goal, offer, channel, source material, and constraints.
3. Ask the assistant to identify missing inputs before producing the final output.
4. Have the assistant follow the skill-specific guidance below.
5. Review the result against the final checklist and ask for revisions where needed.

## Skill-Specific Guidance
- Profound tracked 10,000 prompts across ChatGPT, Perplexity, and Copilot for 14 days.
- The engines generated similar numbers of searches per run but differed greatly in repetition and wording.
- ChatGPT generated highly varied research paths; Perplexity stayed closest to the prompt; Copilot compressed prompts into shorter search-style queries.
- All three often converted broad advice or comparison prompts into concrete fact-retrieval searches.
- Define the prompt family, audience, stage, engine, language, and market.
- Capture observed fan-out queries when the platform or analytics product exposes them.
- Otherwise infer candidate retrieval paths and label them clearly as inferred.
- Decompose the user job into:
- constraints such as price, compatibility, location, and recency
- standards, definitions, reviews, and primary evidence
- Group fan-outs into stable fact families rather than exact query strings.
- Map each family to the current best source and competing cited sources.

## Decision Points And Nuance
The original skill emphasizes: Research basis, Goal, Workflow, Engine considerations, Output.

Use these questions to steer the work:
- What is the intended audience or buyer?
- What source material must be preserved?
- What should the assistant optimize for: clarity, persuasion, accuracy, speed, creativity, or conversion?
- What examples represent the desired quality bar?
- What should the assistant avoid?

## Common Mistakes
- Test across repeated runs and engines; do not expect identical retrieval paths.
- Do not create one doorway page per fan-out query. One strong resource can satisfy several related retrieval needs.
- Required claims, sources, dates, examples, and comparison criteria.

## Copy-And-Paste Prompt
```text
Use the AI retrieval fan-out mapper human guide.

My goal:
[Describe the business outcome]

Audience:
[Describe who this is for]

Context and source material:
[Paste notes, examples, links, or existing copy]

Constraints:
[Tone, length, channel, timeline, must-include items, must-avoid items]

Before producing the final output, ask me for any missing information that would materially improve the result.
```

## Final Checklist
- [ ] The output matches the original goal.
- [ ] The audience and context are reflected in the answer.
- [ ] Important constraints and source material were preserved.
- [ ] The assistant made the relevant decisions explicit.
- [ ] The final artifact is ready to use, review, or hand to the next person.

## Source
This guide was generated from the Profound skill entry for `ai-retrieval-fanout-mapper`.

## Source Skill Notes
These notes preserve the nuance from the original skill. Use them as supporting reference when the workflow above feels too generic.

# AI Retrieval Fan-Out Mapper

## Research basis

This skill operationalizes Profound's research, [What AI engines actually search for and why ChatGPT never searches the same way twice](https://www.tryprofound.com/blog/what-ai-engines-actually-search-for), published April 30, 2026.

Key takeaways from the study:

- Profound tracked 10,000 prompts across ChatGPT, Perplexity, and Copilot for 14 days.
- The engines generated similar numbers of searches per run but differed greatly in repetition and wording.
- ChatGPT generated highly varied research paths; Perplexity stayed closest to the prompt; Copilot compressed prompts into shorter search-style queries.
- All three often converted broad advice or comparison prompts into concrete fact-retrieval searches.

These findings are observational and may change with models and product updates. Use observed fan-outs when available and label inferred fan-outs as hypotheses.

## Goal

Identify the facts, entities, constraints, vocabulary, and supporting evidence an answer engine may retrieve while answering a prompt family.

## Workflow

1. Define the prompt family, audience, stage, engine, language, and market.
2. Capture observed fan-out queries when the platform or analytics product exposes them.
3. Otherwise infer candidate retrieval paths and label them clearly as inferred.
4. Decompose the user job into:
   - entities and category terms
   - factual subquestions
   - comparison criteria
   - constraints such as price, compatibility, location, and recency
   - standards, definitions, reviews, and primary evidence
5. Group fan-outs into stable fact families rather than exact query strings.
6. Map each family to the current best source and competing cited sources.
7. Audit whether the brand has a canonical, current, specific answer.
8. Recommend improving, consolidating, or creating content only when a real information gap exists.
9. Test across repeated runs and engines; do not expect identical retrieval paths.

Do not create one doorway page per fan-out query. One strong resource can satisfy several related retrieval needs.

## Engine considerations

- For ChatGPT, plan for broad vocabulary and adjacent research paths.
- For Perplexity, preserve clear category terms and direct factual matches.
- For Copilot, cover concise search-style formulations and core entities.
- For every engine, verify actual behavior before turning a prior into a rule.

## Output

Return:

1. Prompt-family definition.
2. Fan-out map marking each query as observed or inferred.
3. Fact-family matrix with user need, evidence, current source, competitor source, and gap.
4. Content recommendation identifying which existing page should own each fact.
5. Required claims, sources, dates, examples, and comparison criteria.
6. Test plan by engine with repeated observations and success measures.