majiayu000 Product Requirements Document (PRD)
Overview
A Product Requirements Document (PRD) defines what a product or feature should do and why it matters, without prescribing how to build it. It aligns product, engineering, design, and business stakeholders around a shared understanding of the problem, the proposed solution, success criteria, and scope boundaries.
The PRD sits between the BRD (business justification) and the TRD (technical design):
BRD ("Should we build this?") --> PRD ("What are we building?") --> TRD ("How do we build it?")
Standard PRD Structure
1. Title and Metadata
# PRD: [Feature/Product Name]
| Field | Value |
|----------------|--------------------------------|
| Author | [Name] |
| Status | Draft / In Review / Approved |
| Created | YYYY-MM-DD |
| Last Updated | YYYY-MM-DD |
| Reviewers | [Names] |
| Approvers | [Names] |
| Target Release | [Version / Quarter] |
2. Problem Statement
Clearly articulate the problem from the user's perspective. Avoid solution language.
3. Goals and Non-Goals
Define what success looks like and explicitly state what is out of scope.
4. User Stories
Structured descriptions of desired functionality from the user's perspective.
5. Success Metrics
Quantifiable criteria that determine whether the feature achieved its goals.
6. Requirements
Functional requirements (what the system must do) and non-functional requirements (quality attributes).
7. Milestones and Timeline
Key deliverables and their target dates.
8. Open Questions
Unresolved decisions or unknowns that need answers before or during implementation.
Complete PRD Template
# PRD: [Feature/Product Name]
| Field | Value |
|----------------|--------------------------------|
| Author | [Name] |
| Status | Draft / In Review / Approved |
| Created | YYYY-MM-DD |
| Last Updated | YYYY-MM-DD |
| Reviewers | [Names] |
| Approvers | [Names] |
| Target Release | [Version / Quarter] |
## Problem Statement
[Describe the problem from the user's perspective. What pain point exists?
What opportunity are we addressing? Include data or user research that
substantiates the problem. Avoid mentioning solutions here.]
## Goals
- [Goal 1: measurable outcome]
- [Goal 2: measurable outcome]
- [Goal 3: measurable outcome]
## Non-Goals
- [Explicitly state what is OUT of scope for this effort]
- [Things that might seem related but will not be addressed]
- [Features deferred to future iterations]
## User Stories
### Persona: [Persona Name]
As a [type of user],
I want [action/capability],
So that [benefit/value].
**Acceptance Criteria:**
- [ ] [Criterion 1]
- [ ] [Criterion 2]
- [ ] [Criterion 3]
### Persona: [Persona Name]
As a [type of user],
I want [action/capability],
So that [benefit/value].
**Acceptance Criteria:**
- [ ] [Criterion 1]
- [ ] [Criterion 2]
## Success Metrics
| Metric | Current Baseline | Target | Measurement Method |
|-----------------------|------------------|--------------|--------------------------|
| [Metric 1] | [Current value] | [Target] | [How you will measure] |
| [Metric 2] | [Current value] | [Target] | [How you will measure] |
| [Metric 3] | [Current value] | [Target] | [How you will measure] |
## Functional Requirements
### FR-1: [Requirement Name]
**Priority:** Must Have / Should Have / Could Have / Won't Have
**Description:** [Detailed description of the requirement]
### FR-2: [Requirement Name]
**Priority:** Must Have / Should Have / Could Have / Won't Have
**Description:** [Detailed description of the requirement]
## Non-Functional Requirements
### NFR-1: Performance
[Response time, throughput, latency targets]
### NFR-2: Security
[Authentication, authorization, data protection requirements]
### NFR-3: Scalability
[Expected load, growth projections, scaling requirements]
### NFR-4: Accessibility
[WCAG compliance level, assistive technology support]
### NFR-5: Compatibility
[Browser, device, OS, API version requirements]
## Design and UX
[Link to design mockups, wireframes, or prototypes.
Describe key user flows and interaction patterns.]
## Milestones
| Milestone | Target Date | Description |
|--------------------|-------------|------------------------------------------|
| Design Complete | YYYY-MM-DD | UX mockups approved |
| Alpha | YYYY-MM-DD | Core functionality implemented |
| Beta | YYYY-MM-DD | Feature complete, internal testing |
| GA | YYYY-MM-DD | General availability |
## Dependencies
- [Dependency 1: external service, team, or infrastructure]
- [Dependency 2: other feature or project that must complete first]
## Risks and Mitigations
| Risk | Likelihood | Impact | Mitigation |
|-------------------------------|------------|--------|-------------------------------|
| [Risk 1] | High/Med/Low | High/Med/Low | [Mitigation strategy] |
| [Risk 2] | High/Med/Low | High/Med/Low | [Mitigation strategy] |
## Open Questions
- [ ] [Question 1 -- who owns answering this?]
- [ ] [Question 2 -- target date for resolution?]
- [ ] [Question 3]
Writing Good User Stories
The Standard Format
As a [persona/role],
I want [action or capability],
So that [business value or benefit].
INVEST Criteria for User Stories
| Criterion | Meaning |
|---|---|
| Independent | Can be developed and delivered independently of other stories |
| Negotiable | Details can be discussed and refined collaboratively |
| Valuable | Delivers clear value to the user or business |
| Estimable | Team can estimate the effort required |
| Small | Can be completed within a single sprint |
| Testable | Has clear acceptance criteria that can be verified |
Good vs Poor User Stories
Poor (too vague):
As a user, I want a better search so that I can find things.
Good (specific and valuable):
As a product catalog browser,
I want to filter search results by price range, category, and rating,
So that I can quickly find products that match my budget and preferences.
Acceptance Criteria:
- [ ] Filter panel displays on the search results page
- [ ] Price range filter supports min/max with a slider
- [ ] Category filter shows top-level categories with item counts
- [ ] Rating filter allows selecting minimum star rating (1-5)
- [ ] Filters can be combined (AND logic)
- [ ] Clearing all filters restores the full result set
- [ ] Filter selections persist across pagination
Acceptance Criteria Formats
Checklist format:
Acceptance Criteria:
- [ ] User can upload files up to 50MB
- [ ] Supported formats: PDF, DOCX, XLSX, PNG, JPG
- [ ] Upload progress bar is displayed during upload
- [ ] Error message shown for unsupported file types
Given/When/Then format (for complex scenarios):
Acceptance Criteria:
- Given I am on the search results page
When I set the price filter to $10-$50
Then only products priced between $10 and $50 are displayed
Prioritization Frameworks
MoSCoW Method
| Priority | Meaning | Guideline |
|---|---|---|
| Must Have | Non-negotiable for launch | ~60% of effort |
| Should Have | Important but not critical | ~20% of effort |
| Could Have | Nice to have, low impact if omitted | ~20% of effort |
| Won't Have | Explicitly out of scope for this release | Documented for future |
Example:
## Requirements by Priority
### Must Have
- FR-1: User registration and login
- FR-2: Product search with basic filters
- FR-3: Shopping cart and checkout
- NFR-1: Page load time < 3 seconds
### Should Have
- FR-4: Wishlist functionality
- FR-5: Order tracking dashboard
- NFR-2: Support for 3 languages
### Could Have
- FR-6: Product recommendation engine
- FR-7: Social sharing buttons
### Won't Have (this release)
- FR-8: Marketplace for third-party sellers
- FR-9: Mobile native app
RICE Scoring
| Factor | Description | Scale |
|---|---|---|
| Reach | How many users will this impact per quarter? | Number of users |
| Impact | How much will this impact each user? | 3 = massive, 2 = high, 1 = medium, 0.5 = low, 0.25 = minimal |
| Confidence | How confident are we in estimates? | 100% = high, 80% = medium, 50% = low |
| Effort | How many person-months to complete? | Person-months |
Formula: RICE Score = (Reach x Impact x Confidence) / Effort
| Feature | Reach | Impact | Confidence | Effort | Score |
|---|---|---|---|---|---|
| Search filters | 5000 | 2 | 80% | 2 | 4000 |
| Wishlist | 2000 | 1 | 80% | 1 | 1600 |
| Recommendations | 5000 | 2 | 50% | 4 | 1250 |
| Social sharing | 1000 | 0.5 | 80% | 0.5 | 800 |
Higher RICE score = higher priority.
Relationship to Other Documents
BRD (Business Requirements)
| Answers: "Should we build this? What is the business case?"
| Audience: Leadership, business stakeholders
|
v
PRD (Product Requirements) <-- YOU ARE HERE
| Answers: "What are we building? For whom? What does success look like?"
| Audience: Product, engineering, design
|
+---> ADR (Architecture Decision Records)
| Records: "Why did we choose X over Y?"
|
v
TRD (Technical Requirements)
| Answers: "How do we build it? What are the technical constraints?"
| Audience: Engineering
|
v
Gherkin / Gauge (Executable Specs)
Answers: "How do we verify it works?"
Audience: QA, engineering, product
Best Practices
- Start with the problem, not the solution -- clearly articulate the user pain point before proposing any feature.
- Be specific about non-goals -- explicitly stating what you will not build prevents scope creep and misaligned expectations.
- Make success metrics measurable -- "improve user experience" is not a metric. "Reduce checkout abandonment rate from 68% to 45%" is.
- Write user stories from the user's perspective -- avoid implementation language like "the system shall" in favor of "as a user, I want."
- Include acceptance criteria for every user story -- without testable criteria, it is impossible to know when a story is done.
- Prioritize ruthlessly -- use MoSCoW or RICE to force trade-off decisions early. Not everything can be "Must Have."
- Keep the PRD living -- update the document as requirements evolve. A stale PRD is worse than no PRD because it actively misleads.
- Separate functional from non-functional requirements -- performance, security, and scalability constraints deserve explicit attention.
- Link to related documents -- reference the BRD for business context, ADRs for architectural decisions, and the TRD for technical design.
- Resolve open questions before development starts -- open questions that linger into implementation cause rework and delays.
- Review with all stakeholders -- engineering, design, QA, and product should all review and approve the PRD before implementation begins.