Product
RICE (Reach × Impact × Confidence ÷ Effort)
A prioritisation framework that scores potential product changes by multiplying Reach (how many users affected), Impact (how much per user), and Confidence (how sure you are), divided by Effort (engineering / design weeks).
In plain English
A number you assign to every potential product change so you can rank them objectively. Higher RICE score = ship it sooner.
Example
Feature A: Reach 5,000 users/quarter × Impact 3 (high) × Confidence 0.8 (80%) ÷ Effort 4 weeks = 3,000 RICE score. Feature B: Reach 500 users × Impact 3 × Confidence 0.9 ÷ Effort 1 week = 1,350. Feature A wins despite higher effort.
Formula
RICE = (Reach × Impact × Confidence) / Effort
- Reach: # users affected per quarter
- Impact: 0.25 (minimal) / 0.5 / 1 / 2 / 3 (massive)
- Confidence: 0.5 (low) / 0.8 (medium) / 1.0 (high)
- Effort: person-weeks of work
Why it matters
RICE forces product debates to converge on a number, which beats opinions-driven prioritisation. Teams that use RICE consistently find that 'pet projects' often score badly when scored honestly, and counterintuitive small bets often score well.
Common mistakes
- Gaming Confidence — every PM scores their pet feature 1.0, defeating the purpose. Calibrate against historical accuracy
- Treating RICE as a final decision rather than a discussion starter — strategic context still matters
- Forgetting that low-effort high-reach features stack up — the RICE-optimal roadmap is often '20 small things' rather than '2 big bets'
- Not refreshing scores after new data — RICE scores age