Documentation

PRD (Product Requirements Document)

A written document that describes what a product or feature should do, why it exists, and how success will be measured — before engineering begins.

What is a PRD?

A Product Requirements Document (PRD) is a written artifact that defines the scope, purpose, and success criteria of a product feature or initiative. It is the primary communication tool between product management and engineering, design, and QA.

A well-written PRD answers three questions:

  1. What problem are we solving, and for whom?
  2. What should the product do to solve it?
  3. How will we know if it worked?

What a PRD typically contains

SectionPurpose
Problem statementThe user pain being solved and the evidence for it
Goals & non-goalsWhat success looks like — and what's explicitly out of scope
User storiesScenarios written from the user's perspective
Functional requirementsWhat the system must do (P0 / P1 / P2 priority)
UX & design requirementsAccessibility, layouts, Figma links
Technical constraintsPerformance targets, API dependencies, security
Success metricsMeasurable outcomes with baselines and targets
Timeline & milestonesKey dates for design, engineering, QA, launch
Open questionsUnresolved decisions that could change scope

PRD vs BRD vs MRD

DocumentWho writes itWhat it covers
MRD (Market Requirements Document)Product / MarketingMarket problem, opportunity, customer segments
BRD (Business Requirements Document)Business analyst / PMWhat the business needs — common in enterprise IT
PRD (Product Requirements Document)Product ManagerWhat the product should do — standard in tech companies

In most startups and tech companies, the PRD is the only document that matters. MRDs and BRDs are artifacts of older enterprise processes.


When to write a PRD

Write a full PRD for features that:

  • Span multiple teams or require cross-functional alignment
  • Carry high ambiguity about scope or approach
  • Have significant user journey changes
  • Take more than one sprint to build

For small changes (< 1 sprint, single team, low ambiguity), a well-written Jira or Linear ticket with acceptance criteria is usually enough.


Common PRD mistakes

  • Starting with solutions instead of problems. A PRD that leads with "build a dashboard" instead of "users can't see their data without exporting to Excel" is a requirements document for the wrong thing.
  • No non-goals. Listing what you're not building is as important as what you are. Without it, scope creep is inevitable.
  • Unmeasurable success metrics. "Improve user experience" is not a metric. "Reduce time-to-complete onboarding from 8 minutes to 4 minutes" is.
  • Written in isolation. The best PRDs are co-authored with an engineering lead. Their early input on feasibility prevents late rewrites.

Frequently asked questions

How long should a PRD be?

2–5 pages for most features. 8–12 pages for large platform bets. If it's longer than 15 pages, split it. The goal is to be unambiguous, not comprehensive.

Should engineering be involved before the PRD is written?

Yes. Involve an engineering lead in the problem statement and scope sections. Their input on feasibility and constraints makes requirements more realistic and avoids late rewrites.

Does every feature need a PRD?

No. For changes under 1 sprint with low ambiguity, a detailed Jira ticket is sufficient. Reserve the full PRD for features with cross-team scope, significant user journey changes, or high uncertainty.

What's the difference between a PRD and a spec?

A PRD defines what to build and why (product perspective). A spec defines how to build it (engineering perspective). A PRD precedes the spec. In some teams, the spec is written inside the PRD as a separate section.

Apply PRD to your real product data

PMRead ingests customer feedback, interviews, and Slack threads — and generates PRDs grounded in real evidence.

Try PMRead free →