SKILL FILE

Jobs-to-be-Done Extractor with AI

Turn raw customer research into Jobs-to-be-Done statements that reveal what users are really trying to accomplish — beyond surface-level feature requests.

Download Skill File ↓

What this skill file teaches Claude

Drop one markdown file into your repo. Claude Code learns how to run this entire workflow.

1

JTBD statement extraction

Properly formatted "When [situation], I want to [action], so I can [outcome]" statements pulled from interview transcripts, survey responses, or support tickets.

2

Functional / emotional / social jobs

Separates jobs into three layers so you see what users do, how they want to feel, and what they want others to think.

3

Opportunity scoring

Importance + (Importance - Satisfaction) to show which jobs are underserved — where you can win.

4

Feature → need translation

Maps what users asked for to what they actually need. Stops you building the wrong thing politely.

5

Evidence tracking

Every JTBD statement linked back to the quote or behaviour that supports it. No invented insights.

What you can build with this

Reframe feature requests

Take 20 customer feature requests and surface the 3 underlying jobs they're really asking you to solve.

Find innovation opportunities

In a saturated market, the unmet jobs are where the next wave of products will be built.

Train your team to think in jobs

Onboard new PMs and designers on how to interview for jobs instead of features.

Get the full skill file

Everything above is 80% of the skill file. Download the complete version with full implementation details, agent prompts, and ready-to-run scripts.

Common questions

A framework that defines products in terms of the "job" a user "hires" them to do. Instead of "users want a calendar feature", JTBD says "when planning my week, I want to see all my commitments in one place, so I can decide what I have time for." It reframes features as means to ends.
Ideally yes — interview transcripts, survey responses, support tickets, or recorded sales calls all work. The richer the source material, the better the extraction. You can also use the skill with anonymised internal feedback if customer data is scarce.
Personas describe who the user IS (demographics, behaviours, goals). JTBD describes what the user is TRYING TO DO. Both are useful — JTBD tends to be more stable over time and across user segments, because the underlying job rarely changes even when the user does.
Usually 3–7 functional jobs, plus emotional and social jobs layered on top. Too few and you've oversimplified; too many and you haven't found the patterns. A single user interview typically surfaces 5–10 candidate jobs.
No — it depends on research. JTBD is a synthesis framework that organises the research you've already done. You still need interviews, surveys, or behavioural data to feed it.

Ready to automate?

SoloStack gives you every skill pre-installed — scraping, marketing, sales, CRM, and more. One repo. Every department.

Book a Call →