← Files AIsa GTMARCHIVED FILE

skills/customer-research/README.md

1.97 KB · Oct 5, 2026 · 18:25 UTC

↓ Download file

# Customer Research

Customer Research turns interviews, surveys, support material and public customer conversations into source-backed voice-of-customer insight. It helps teams understand what customers are trying to do, what triggers action, what hurts, which alternatives they use and which language they use to describe the problem.

## What it can deliver

- evidence ledger with dates, source type and short verbatim excerpts;
- top themes and confidence based on independent evidence;
- JTBD, pains, triggers, desired outcomes, alternatives and objections;
- segment or Persona hypotheses, with a reliable Persona only when the evidence threshold is met;
- implications for positioning, product, content or sales research;
- contradictions, sample bias, unknowns and follow-up interview/survey questions.

## Example requests

> Synthesize these 12 customer interviews into the top jobs, pains and objections. Keep direct quotes and distinguish facts from hypotheses.

> Research what freelance designers say about client approval tools in English-speaking markets, using a small Reddit sample and citing each source.

> We only have a product name. Propose a provisional customer segment, research scope and the questions we should validate before building a Persona.

## Research modes and boundaries

Supplied material is analyzed first and prevents duplicate external research. Public research uses a small, explicitly scoped sample of available AIsa MCP capabilities; paid calls are quoted and require authorization before execution. Empty, blocked or contradictory sources remain unknown and are never turned into negative evidence.

This skill researches customers and their needs; it does not build prospect lists, compare competitors, audit a website, send surveys or outreach, change CRM, or make individual decisions. Public conversations are self-selected evidence, not representative population samples. Do not infer sensitive attributes or fabricate customers, participants, quotes or outcomes.

SHA-256: be54d9b95101db98bad08d3d28b90c439a3fc23e35b5dd165542225f6b06ddc1