Requirements Prioritization

Who this is for

This lesson is for Business Analysts who gather requirements and must decide what to build first. It helps you drive alignment between stakeholders, developers, and users under limited time and budget.

Prerequisites

• Basic understanding of requirements (user stories, epics, acceptance criteria)
• Ability to estimate effort with your team at a rough level (S/M/L or story points)
• Familiarity with stakeholder goals and key business outcomes

Why this matters

Real BA tasks where prioritization is critical:

• Choosing the Minimal Viable Release from a long list of requests
• Sequencing compliance and risk items before an audit deadline
• Balancing quick wins vs. strategic bets for quarterly planning
• Resolving conflicts when two departments want different features first
• Keeping the roadmap realistic within capacity constraints

Concept explained simply

Prioritization is deciding the order and grouping of requirements to maximize value and reduce risk under constraints (time, budget, dependencies).

Mental model

Think of your backlog as a queue at a clinic. Patients (items) differ on urgency, impact, and effort. Your job is to seat the most impactful, urgent, safe-to-handle cases first, while not starving critical but slower procedures. Use clear rules, not loud voices.

• Value: Who benefits and how much?
• Risk/Urgency: What happens if we delay?
• Effort: How long and how complex?
• Confidence: How sure are we about value and effort?
• Constraints: Deadlines, dependencies, regulatory must-haves

Core frameworks you can use

• MoSCoW: Must / Should / Could / Won’t (for this release)
• Value vs. Effort matrix: Quick wins first (High Value, Low Effort)
• RICE: (Reach × Impact × Confidence) / Effort
• WSJF (Scaled Agile): (Cost of Delay) / Duration
• Kano: Must-haves, Performance features, Delighters (user satisfaction lens)
• Risk/Compliance flagging: Items that prevent fines or outages jump the queue

Worked examples

Example 1: MoSCoW for an internal tool

Context: Operations team needs a dashboard before quarter-end.

• Must: Authentication, Role-based view, Core KPIs
• Should: Export to CSV, Data filters
• Could: Dark mode, Custom widgets
• Won’t (this release): Mobile app

Outcome: A usable dashboard ships sooner; nice-to-haves are scheduled later.

Example 2: RICE for product features

• A: Self-serve password reset — Reach 8k/month, Impact 2 (on a 0.25–3 scale), Confidence 0.9, Effort 5 → RICE = (8000 × 2 × 0.9) / 5 = 2880
• B: Advanced search — Reach 5k, Impact 2, Confidence 0.7, Effort 8 → 875
• C: Billing alerts — Reach 3k, Impact 3, Confidence 0.8, Effort 3 → 2400

Order: A (2880), C (2400), B (875). You ship reset and billing alerts before advanced search.

Example 3: WSJF for platform backlog

Scores (higher is better):

• Reduce page load time: Cost of Delay 36, Duration 3 → WSJF 12
• New reporting API: CoD 24, Duration 4 → WSJF 6
• Security patch: CoD 40, Duration 10 → WSJF 4

Order by WSJF: Performance (12), API (6), then Patch (4). If the patch is compliance-critical, flag it as Must and pull it earlier despite the score.

Step-by-step prioritization recipe

1. List and clarify items (one outcome per item)
2. Mark constraints: deadlines, dependencies, regulatory
3. Select a method (MoSCoW, RICE, WSJF, etc.)
4. Get rough inputs with the team (effort, reach, impact, etc.)
5. Score or categorize; sort the list
6. Review conflicts; explicitly handle dependencies and compliance
7. Slice into releasable increments (MVP, MMP)
8. Create a one-page rationale: method used, inputs, assumptions, resulting order
9. Share, get buy-in, and lock the scope for this iteration

Interactive: Exercise 1 — Prioritize with RICE

Use the items and inputs below. Calculate RICE = (Reach × Impact × Confidence) / Effort. Impact: 0.25, 0.5, 1, 2, 3 scale. Confidence as a decimal (e.g., 0.8). Effort in person-weeks.

• Item A — In-app onboarding: Reach 4000, Impact 2, Confidence 0.7, Effort 6
• Item B — Two-factor auth: Reach 3000, Impact 3, Confidence 0.9, Effort 5
• Item C — Team invites: Reach 5000, Impact 1, Confidence 0.8, Effort 4
• Item D — Usage digest email: Reach 7000, Impact 0.5, Confidence 0.9, Effort 2

Task:

• Compute RICE for A–D
• Sort descending
• Identify one assumption that most affects the order and how you’d validate it

Note: Everyone can do the exercise and the test. Progress is saved only for logged-in users.

Self-check checklist

• I used an explicit method (MoSCoW/RICE/WSJF) and stated assumptions
• Compliance and hard deadlines are clearly marked
• Dependencies are resolved or sequenced
• We capped Must-haves and produced a release-sized slice
• Confidence is recorded; follow-ups are planned for low-confidence items

Common mistakes and how to self-check

• Everything is “Priority 1”: Force a cap and show trade-offs versus team capacity
• Ignoring dependencies: Add a dependency check step before finalizing order
• Scoring theater: Numbers without meaning; keep criteria simple and visible
• Underrating risk/compliance: Overlay a “must not fail” flag for audits, security
• No record of why: Store a one-page rationale to cut rework in stakeholder meetings

Practical projects

• Value vs. Effort board: Place 12 backlog items into the 2×2 and publish the resulting top 5
• RICE mini-roadmap: Score 8 features and propose a two-release plan with rationale
• Risk-first slice: Identify regulatory items and craft a minimal compliance-ready release

Learning path

• Listen and clarify: Elicitation interviews, observation, and notes
• Structure: Break epics into testable, independent stories
• Prioritize: Apply MoSCoW/RICE/WSJF and handle constraints
• Plan: Create release slices and acceptance criteria
• Validate: Pilot, feedback loops, and re-prioritize

Next steps

• Pick one method and use it on your current backlog this week
• Share a one-page prioritization rationale with stakeholders
• Run the quick test below to confirm understanding

Mini challenge

You have 10 items and only 6 weeks of capacity. Two items are audit-critical. Propose an MVP that ships in 4 weeks, includes audit items, and leaves a 2-week buffer for one high-value bet. Write your top 5 items and the reason each made the cut.

Quick Test

Take the test below. Everyone can attempt it; progress saving requires login.