Feature Discovery

Discover and validate feature ideas in 4 steps: problem exploration, solution ideation, validation plan, and specification.

Estimated time: Varies with inputs and review

By the end of this workflow you'll have a defined problem, three solution concepts, a validation plan with success criteria, and a draft feature spec.

  1. Problem Exploration — Deep-dive into the problem space with customer evidence and impact analysis.
  2. Solution Ideation — Generate and evaluate multiple solution approaches for the problem.
  3. Validation Plan — Design experiments and research to validate the proposed solution before building.
  4. Feature Specification — Write the feature spec with requirements, acceptance criteria, and launch plan.

Work through the steps

4 stepsresearch
Progress0 of 4 completed
Step 1 of 4: Problem Exploration
Deep-dive into the problem space with customer evidence and impact analysis.
Prompt Preview
Use supplied facts and sources. Label estimates and proposed plans. Do not invent quotes, results, customer evidence, or citations. Mark missing facts [NEEDS INPUT]; assumptions from earlier steps remain assumptions.

You are a product discovery expert helping a PM explore a problem space.

## Task
Conduct a deep problem exploration for: **[Problem Area]**

## Customer Evidence Available
[Customer Evidence]

## Deliverables

### 1. Problem Definition
- **Problem statement**: Clear, specific, measurable
- **Who is affected**: User segments, personas, and their context
- **Frequency**: How often is this problem encountered?
- **Severity**: How painful is it when encountered? (1-10 with criteria)
- **Current workarounds**: What do users do today?

### 2. Jobs-to-Be-Done Analysis
Identify 3-5 jobs users are trying to accomplish:
- **Job statement**: When [situation], I want to [motivation], so I can [outcome]
- **Current solutions**: How they address this job today
- **Satisfaction level**: How well current solutions work (1-10)
- **Switching triggers**: What would motivate them to try something new

### 3. Impact Analysis
- **Business impact**: Revenue at risk, support costs, churn attribution
- **User impact**: Time wasted, frustration, abandoned workflows
- **Opportunity cost**: What users cannot do because of this problem
- **Cascade effects**: How this problem affects other metrics and experiences

### 4. Evidence Gaps
What do you NOT know yet?
- Questions that need answers before solution design
- Assumptions that need validation
- Data that would change your prioritization
- Customer segments you have not heard from

### 5. Problem Prioritization Score
Rate on 1-10:
- **Frequency**: How often does this occur?
- **Severity**: How painful when it occurs?
- **Breadth**: What percentage of users are affected?
- **Trend**: Is this getting better or worse?
- **Strategic fit**: How well does solving this align with strategy?

Calculated priority score with confidence level.

## Evidence and accuracy
- Use supplied facts and verified sources. Do not invent metrics, quotes, people, company details, commitments, or personal experience.
- Mark assumptions [ASSUMPTION], estimates [ESTIMATE: method], and unresolved questions [UNCERTAIN: reason]. Leave unavailable values unfilled rather than guessing.
- Explain confidence as high, medium, or low using the evidence available. These labels are judgments, not calibrated probabilities. Give numerical probabilities only when a stated method supports them.
- Treat preset weights, scores, timelines, and targets in this template as starting examples. Adapt them to the task and explain changes; they are not universal benchmarks or approved commitments.
- Define score scales and directions before calculating totals. Keep units, denominators, and time periods consistent. Do not average away a critical blocker.
- Use only exact supplied or verified quotes with attribution. Label requested fictional examples as illustrative. Distinguish observed behavior from inferred motives or causes.
- Use only the sections and rows the task needs. Write plainly, preserve necessary technical terms, and avoid unsupported benefits or forced specificity.
- Identify missing information needed for a decision. Recommendations remain proposals until reviewed by the responsible team.
AI Response
Paste the AI's response here after running the prompt
0 words