About product-discovery
product-discovery is a user research and problem-validation skill for product teams. It covers interviews, JTBD, opportunity assessment, and insight synthesis, with an emphasis on validating problems before solutions and maintaining ongoing discovery.
Key features
- 01User interview and survey design
- 02JTBD problem exploration
- 03Opportunity scoring and prioritization
- 04Research synthesis and segmentation
- 05Prototype and usability testing
Use cases
- 01Explore customers’ unarticulated problems
- 02Validate whether a specific pain is worth solving
- 03Prioritize a product roadmap by opportunity
- 04Test prototypes and assess solutions
Install
Copy the command and run it in a terminal.
npx skills add https:/ / github.com/ ncklrs/ startup-os-skills --skill product-discoveryAfter installing, type/product-discoveryin Claude Code to confirm it loaded.
Bundled files
References and scripts installed along with SKILL.md.
rules/_sections.mdanalysis-competitive.mdanalysis-opportunity.mdanalysis-segmentation.mdanalysis-synthesis.mddiscovery-jtbd.mddiscovery-problems.mdresearch-interviews.mdresearch-surveys.mdtesting-prototype.mdtesting-usability.mdDetails
From the project READMEProduct Discovery
Strategic user research and problem validation expertise — from interview techniques and JTBD to opportunity sizing and insight synthesis.
Philosophy
Great products start with great problems. Discovery is how you find problems worth solving for people who will pay.
The best product discovery:
- Talk to users, not stakeholders — Customers know their problems, not solutions
- Validate problems before solutions — Build the right thing, then build it right
- Quantify and qualify — Numbers tell you what, conversations tell you why
- Continuous over batched — Weekly habits beat quarterly projects
How This Skill Works
When invoked, apply the guidelines in rules/ organized by:
research-*— User interview techniques, survey design, research opsdiscovery-*— Problem discovery, JTBD framework, validationanalysis-*— Synthesis, segmentation, competitive analysistesting-*— Prototype testing, usability testing
Core Frameworks
Discovery Process
| Phase | Activities | Outputs |
|---|---|---|
| Explore | Interviews, observation, data mining | Problem space map |
| Validate | Problem interviews, surveys, experiments | Validated problems |
| Prioritize | Opportunity scoring, segmentation | Prioritized roadmap |
| Test | Prototype testing, usability studies | Solution validation |
Jobs-to-be-Done Framework
┌─────────────────────┐
│ FUNCTIONAL JOB │
│ (What they do) │
└──────────┬──────────┘
│
┌────────────────┼────────────────┐
│ │ │
▼ ▼ ▼
┌──────────┐ ┌──────────┐ ┌──────────┐
│ EMOTIONAL│ │ SOCIAL │ │ CONTEXT │
│ JOB │ │ JOB │ │ (When/ │
│ (Feel) │ │ (Appear) │ │ Where) │
└──────────┘ └──────────┘ └──────────┘
Opportunity Scoring (OST)
| Factor | Weight | Description |
|---|---|---|
| Importance | 40% | How important is this job to the customer? |
| Satisfaction | 30% | How satisfied are they with current solutions? |
| Frequency | 20% | How often do they encounter this problem? |
| Willingness to Pay | 10% | Will they pay to solve this? |
Opportunity Score = Importance + max(Importance - Satisfaction, 0)
Research Method Selection
┌─────────────────────────────────────────────────────────────┐
│ GENERATIVE RESEARCH │
│ (Discover unknown unknowns) │
│ ┌───────────┐ ┌───────────┐ ┌───────────┐ │
│ │Contextual │ │ Discovery │ │ Diary │ │
│ │ Inquiry │ │ Interviews│ │ Studies │ │
│ └───────────┘ └───────────┘ └───────────┘ │
├─────────────────────────────────────────────────────────────┤
│ EVALUATIVE RESEARCH │
│ (Validate known hypotheses) │
│ ┌───────────┐ ┌───────────┐ ┌───────────┐ │
│ │ Usability │ │ A/B │ │ Prototype │ │
│ │ Testing │ │ Testing │ │ Testing │ │
│ └───────────┘ └───────────┘ └───────────┘ │
├─────────────────────────────────────────────────────────────┤
│ QUANTITATIVE RESEARCH │
│ (Measure and prioritize) │
│ ┌───────────┐ ┌───────────┐ ┌───────────┐ │
│ │ Surveys │ │ Analytics │ │ Card │ │
│ │ │ │ Review │ │ Sorting │ │
│ └───────────┘ └───────────┘ └───────────┘ │
└─────────────────────────────────────────────────────────────┘
Customer Segmentation Matrix
| Dimension | Consumer (B2C) | Business (B2B) |
|---|---|---|
| Demographics | Age, income, location | Company size, industry, revenue |
| Behavior | Usage patterns, purchase history | Buying process, tech stack |
| Psychographics | Values, lifestyle, attitudes | Company culture, risk tolerance |
| Needs | Problems, goals, aspirations | Business outcomes, KPIs |
Continuous Discovery Cadence
Weekly:
├── 2-3 customer interviews
├── Review analytics/feedback
└── Update opportunity backlog
Monthly:
├── Synthesis session
├── Prioritization review
└── Stakeholder alignment
Quarterly:
├── Deep-dive research sprint
├── Competitive analysis refresh
└── Segment review
Interview Quick Reference
| Interview Type | When to Use | Key Questions |
|---|---|---|
| Discovery | Exploring problem space | "Tell me about the last time..." |
| Problem | Validating specific pain | "How painful is this 1-10? Why?" |
| Solution | Testing concepts | "Would this solve your problem?" |
| JTBD | Understanding motivation | "What were you trying to accomplish?" |
| Usability | Testing interfaces | "What do you expect to happen?" |
Anti-Patterns
- Solution-first discovery — Falling in love with solutions before validating problems
- Leading the witness — Asking questions that suggest desired answers
- Confirmation bias — Only hearing what supports your hypothesis
- Sample of one — Making decisions from a single interview
- Proxy research — Asking salespeople instead of customers
- Feature requests as research — Users ask for features, not problems
- Analysis paralysis — Researching forever, never deciding
- HiPPO-driven — Highest Paid Person's Opinion overriding data