skills.rest
spikelab
spikelabCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

pm-jtbd-synthesis

Synthesize Jobs-to-be-Done and Forces of Progress from customer interview transcripts.

Updated May 3, 2026
Search Tags:#synthesis#product discovery#jobs-to-be-done#customer interviews#jtbd#job stories#forces of progress
One-click install
npx skills add https://github.com/spikelab/multiplai-cc-mktplace --skill pm-jtbd-synthesis-spikelab
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pm-jtbd-synthesis
Source: https://github.com/spikelab/multiplai-cc-mktplace/tree/main/plugins/multiplai-pm/skills/pm-jtbd-synthesis
Command: npx skills add https://github.com/spikelab/multiplai-cc-mktplace --skill pm-jtbd-synthesis-spikelab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Product teams collect customer interview transcripts but struggle to turn them into structured, verifiable discovery artifacts. This Skill converts raw transcripts into Jobs-to-be-Done synthesis with line-number-anchored evidence, so product decisions rest on traceable customer signal rather than memory or opinion. ## Core Features & Use Cases - Forces of Progress Extraction: Extracts push, pull, anxiety, and habit forces from each transcript with verbatim quotes, line-number anchors, speaker attribution, and STRONG/SUPPORTED/WEAK confidence tags. - Cross-Transcript Job Clustering: Clusters functional and emotional/social jobs across multiple transcripts, caps output at 7 clusters, and surfaces tensions and single-transcript outliers. - Three Dated Artifacts: Emits a full synthesis report, canonical When/I-want/So-I-can job stories, and an Opportunity-Solution Tree stub into an INBOX folder. - Use Case: Point the Skill at a folder of five customer discovery call transcripts and receive a JTBD synthesis report with anchored quotes, clustered jobs, and an OST stub ready for the product team to fill in solutions. ## Quick Start Synthesize the customer interview transcripts in the ./interviews folder into a JTBD report with job stories and an opportunity-solution tree stub.

Frequently Asked Questions about pm-jtbd-synthesis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I synthesize customer interview transcripts into Jobs-to-be-Done?▼

Provide a file path, folder, or comma-separated list of transcript files as the source argument. The skill runs three passes: per-transcript forces extraction, cross-transcript job clustering, and artifact emission, producing a synthesis report, job stories, and an OST stub.

What are the Forces of Progress in JTBD analysis?▼

The Forces of Progress are push (dissatisfaction with the current way), pull (attraction to a new solution), anxiety (fear of switching), and habit (inertia of the present). The skill extracts each force with verbatim quotes, line numbers, and confidence ratings.

Can I use audio files instead of text transcripts?▼

Yes, pass --audio true and the skill routes audio files through the transcribe skill from the multiplai-media plugin first. If that plugin is not installed, you must provide text transcripts instead.

What output files does the JTBD synthesis produce?▼

It writes three date-stamped files to ./INBOX/ if it exists: a full synthesis report, job stories in When/I-want/So-I-can format, and an Opportunity-Solution Tree stub. With --depth quick, only the report with an embedded job-stories section is emitted.

Why does the skill require line numbers in transcripts?▼

Line numbers anchor every force, job cluster, and quote to a verifiable source location, making the synthesis auditable. If a source lacks line numbers, the skill notes in Method Notes that quotes are unanchored.

What are the limitations of JTBD synthesis with few transcripts?▼

With thin input such as three transcripts, clusters cannot be called high-confidence and the skill states this explicitly. Single-transcript jobs are held as outliers rather than promoted, and confidence tags are never inflated from WEAK to STRONG.

Related Skills

Explore More Product & Management Skills→
EstomEstom

transcribe

Transcribe audio files into text with optional speaker diarization.

Community
Intermediate
AbsolutelySkilledAbsolutelySkilled
207

product-discovery

Analyze product discovery workflows to map opportunities and design falsifiable experiments.

Official
Advanced
frumu-aifrumu-ai
114

product-user-research

Synthesize qualitative and quantitative user research data into structured insights.

Official
Intermediate
rnw-communityrnw-community
78

brainstorming

Guide users through design exploration and approval before implementation.

Official
Intermediate