---
name: product-market-fit
version: 1.0.0
category: Product & Strategy
domain: product-market-fit
author: Matt Warren
license: MIT
status: production
updated: 2026-02-07
activation_triggers:
  - "product market fit"
  - "PMF"
  - "Sean Ellis"
  - "do people want this"
  - "validate my idea"
  - "customer validation"
  - "retention analysis"
  - "pivot"
tools: []
---

# Product-Market Fit

PMF surveys, Sean Ellis test, retention analysis, and pivot frameworks.

## Purpose

Determine whether you've found product-market fit and, if not, what to change. PMF is the single most important milestone for any startup.

## Workflow

### Step 1: Gather Context
- Product/service description
- Current customer count and engagement
- How customers find you today
- Retention/repeat usage data (if available)
- What customers say they love (and hate)

### Step 2: Sean Ellis Test Design
Survey question: "How would you feel if you could no longer use [product]?"
- Very disappointed
- Somewhat disappointed
- Not disappointed
- N/A — I no longer use it

**Benchmark:** 40%+ "Very disappointed" = PMF signal

Design the full survey (5-8 questions) to understand:
- Who are the most passionate users?
- What's the primary benefit they get?
- What would they use as an alternative?
- How did they discover you?

### Step 3: PMF Assessment
Based on data provided:
- **Strong PMF signals:** High retention, word-of-mouth growth, pull from customers
- **Weak PMF signals:** High churn, feature requests that change the core, price sensitivity
- **No PMF signals:** Growth only from paid acquisition, low engagement, high support volume

### Step 4: Pivot Framework (if needed)
If PMF isn't there, evaluate:
- **Zoom in:** Double down on the feature users love most
- **Zoom out:** Your feature should be a platform
- **Customer pivot:** Same product, different audience
- **Need pivot:** Same audience, different problem
- **Channel pivot:** Same product, different distribution

### Step 5: Action Plan
Concrete next steps based on assessment.

## Output Format
```markdown
## PMF Assessment: [Product]

### Current Signals
| Signal | Status | Evidence |
|--------|--------|----------|
| Retention | [Strong/Weak] | [Data] |
| Word of mouth | [Strong/Weak] | [Data] |

### Sean Ellis Survey
[Survey questions]

### Assessment
[PMF status and reasoning]

### Recommended Actions
1. [Action]
2. [Action]
```

## Constraints
- PMF is not binary — present it as a spectrum
- Don't declare PMF based on vanity metrics (signups, downloads)
- Be honest if the data suggests a pivot is needed
- Note that PMF can be lost — it's not permanent