Home
cd ../playbooks
Product ManagementIntermediate

PRD Development

Build a structured PRD that connects problem, users, solution, and success criteria across 8 phases — from scattered Slack threads to engineering-ready document for major initiatives.

2-3 days
By communitySource
#prd#requirements#documentation#alignment#discovery

Your team is two weeks into building, and engineering is asking "what about international users?" while design is asking "is this dismissable?" and the CEO is asking "why this and not the other thing?" That's not an engineering problem — it's a missing PRD. A real PRD answers seven questions before code starts and saves you the rebuild.

Who it's for: PMs writing PRDs for major features, product leads aligning cross-functional teams, founders documenting initiatives, engineering managers needing handoff context

Example

"Write a PRD for guided onboarding to fix our 60% drop-off" → 8-phase walkthrough: exec summary, problem (60% drop-off + customer quotes), personas (Solo Sam), strategy (Q1 churn OKR), solution (3-step checklist), metrics (40%→60% activation), stories with acceptance criteria, out of scope

CLAUDE.md Template

New here? 3-minute setup guide → | Already set up? Copy the template below.

# PRD Development

Build a structured PRD (Product Requirements Document) connecting problem, users, solution, and success criteria. Move from scattered Slack threads to a clear, comprehensive document that aligns stakeholders and provides engineering context.

Not a waterfall spec. A living document that captures strategic context, customer problems, proposed solutions, and metrics — and evolves as you learn through delivery.

## What a PRD Answers

1. What problem are we solving?
2. For whom?
3. Why now?
4. What are we building?
5. How will we measure success?
6. What are the requirements?
7. What are we NOT building?

## Standard Structure

```markdown
# [Feature/Product Name] PRD

## 1. Executive Summary
One-paragraph: problem + solution + impact

## 2. Problem Statement
Who, what, why painful, evidence

## 3. Target Users & Personas
Primary + secondary personas, JTBD

## 4. Strategic Context
Business goals (OKRs), market opportunity, competitive context, why now

## 5. Solution Overview
High-level description, user flows, key features

## 6. Success Metrics
Primary, secondary, guardrails

## 7. User Stories & Requirements
Epic hypothesis, stories with acceptance criteria, edge cases

## 8. Out of Scope
What we're NOT building and why

## 9. Dependencies & Risks
Technical, external, team dependencies + mitigations

## 10. Open Questions
Unresolved decisions
```

## When to Use

**Use:** Major feature/product initiative, cross-functional alignment needed, documenting decisions for future reference, onboarding new team members.
**Don't use:** Small bug fixes, problem already aligned (just write stories), continuous discovery experiments (use Lean UX Canvas).

## Application: 8 Phases over 2-4 Days

### Phase 1: Executive Summary (30 min)

Format: "We're building [solution] for [persona] to solve [problem], which will result in [impact]."

Write first (forces clarity), refine last.

### Phase 2: Problem Statement (60 min)

Use `problem-statement`. Include:
- **Who:** Specific persona (not "users")
- **What:** The actual problem (not solution disguise)
- **Why painful:** User + business impact
- **Evidence:** Customer quotes, analytics, support tickets

### Phase 3: Target Users & Personas (30 min)

Use `proto-persona`. Document primary + secondary personas: role, company size, tech savvy, goals, pains, current behavior.

### Phase 4: Strategic Context (45 min)

- **Business goals** linked to OKRs
- **Market opportunity** — `tam-sam-som-calculator` (optional)
- **Competitive landscape**
- **Why now** — urgency rationale

### Phase 5: Solution Overview (60 min)

High-level (2-3 paragraphs), user flows, key features. Optional: wireframes, story map (`user-story-mapping-workshop`).

Keep high-level. Let design own UI details.

### Phase 6: Success Metrics (30 min)

- **Primary metric** — the ONE you optimize for (current → target, timeline)
- **Secondary metrics** — additional indicators
- **Guardrail metrics** — what must NOT regress

### Phase 7: User Stories & Requirements (90-120 min)

Use `epic-hypothesis`, `epic-breakdown-advisor`, `user-story`. Output: hypothesis + 3-10 stories with acceptance criteria + constraints/edge cases.

### Phase 8: Out of Scope & Dependencies (30 min)

- **Out of scope** with rationale
- **Dependencies** — design, engineering, external
- **Risks + mitigations**
- **Open questions**

## Workflow Summary

```
Day 1: Phases 1-4 (summary, problem, personas, strategy)
Day 2: Phases 5-7 (solution, metrics, stories)
Day 3: Phase 8 + review/refine
Day 4 (optional): Stakeholder review
```

**Time investment:** Fast track 1.5-2 days; typical 2-3 days; complex 3-4 days.

## Common Pitfalls

1. **PRD written in isolation** → collaborate on stories with design + eng before finalizing
2. **No evidence in problem statement** → quotes, analytics, tickets required
3. **Solution too prescriptive (pixel-level UI)** → keep high-level; let design own details
4. **No success metrics** → always define primary metric
5. **Out of scope undocumented** → explicit list prevents scope creep

## References

**Phase 2:** `problem-statement`, `problem-framing-canvas`, `customer-journey-mapping-workshop`
**Phase 3:** `proto-persona`, `jobs-to-be-done`
**Phase 4:** `tam-sam-som-calculator`
**Phase 5:** `user-story-mapping-workshop`
**Phase 7:** `epic-hypothesis`, `epic-breakdown-advisor`, `user-story`

**External:**
- Martin Eriksson, *How to Write a Good PRD* (2012)
- Marty Cagan, *Inspired* (2017)
- Amazon, *Working Backwards* (PR/FAQ alternative)
README.md

What This Does

Orchestrates 8 phases of PRD creation — exec summary, problem statement, personas, strategic context, solution overview, success metrics, user stories, out of scope — over 2-4 days. Calls component skills (problem-statement, proto-persona, epic-hypothesis, user-story) at each phase.

Anti-pattern: PRD as waterfall spec. This is a living document for alignment, not a frozen contract.


Quick Start

mkdir -p ~/Documents/PRD
mv ~/Downloads/CLAUDE.md ~/Documents/PRD/
cd ~/Documents/PRD
claude

Provide discovery insights, persona research, business goals, and competitive context. Claude orchestrates the 8-phase workflow and outputs a complete PRD.


The 8 Phases

Phase Time Output
1. Executive Summary 30 min One paragraph (problem + solution + impact)
2. Problem Statement 60 min Who/what/why/evidence
3. Target Users 30 min Primary + secondary personas
4. Strategic Context 45 min OKRs, market, competition, why now
5. Solution Overview 60 min High-level description (not pixel spec)
6. Success Metrics 30 min Primary + secondary + guardrails
7. User Stories 90-120 min Epic hypothesis + stories with acceptance criteria
8. Scope & Dependencies 30 min Out of scope, dependencies, risks, open questions

Tips & Best Practices

  • Write the exec summary first, refine it last. Forces clarity upfront, captures lessons at the end.
  • Evidence in the problem statement. Customer quotes + analytics + support tickets — not "we believe."
  • Solution overview is high-level. Don't spec UI in the PRD. Let design own pixels.
  • Always define a primary metric. Without it, you can't know if the feature succeeded.
  • Document out of scope explicitly. Prevents scope creep and stakeholder surprise.

Common Pitfalls

  • PRD written in isolation, presented as finished doc (no buy-in)
  • "We believe users have this problem" with no data
  • Pixel-level UI spec that removes design collaboration
  • No success metrics defined upfront
  • No "out of scope" section, leading to scope creep

$Related Playbooks