Feature Prioritization RICE Score

Formula

RICE Score = (Reach × Impact × Confidence%) / Effort

Reach is users affected per quarter. Impact is 0.25-3 scale. Confidence is 0-100% certainty. Effort is person-weeks. Higher scores indicate higher priority features.

Worked Examples

Example 1: E-commerce Search Improvement

Problem:Reach: 50,000 users/quarter (all searchers). Impact: 2 (High - doubles conversion). Confidence: 70% (A/B test suggested). Effort: 4 person-weeks.

Solution:RICE = (Reach × Impact × Confidence) / Effort RICE = (50,000 × 2 × 0.70) / 4 RICE = 70,000 / 4 RICE = 17,500 This is a very high score because: - Large reach (50K users) - High impact (2x conversion) - Decent confidence (70%) - Moderate effort (4 weeks) Recommendation: High priority, schedule for next sprint.

Result:RICE: 17,500 | Critical Priority | Ship next sprint

Example 2: Mobile Push Notifications

Problem:Reach: 20,000 (mobile users). Impact: 0.5 (Low - minor engagement bump). Confidence: 80%. Effort: 2 person-weeks.

Solution:RICE = (20,000 × 0.5 × 0.80) / 2 RICE = 8,000 / 2 RICE = 4,000 Moderate score because: - Good reach - Low impact (only 0.5) - High confidence - Low effort Effort-adjusted: Good ROI despite low impact. Consider if mobile engagement is strategic priority.

Result:RICE: 4,000 | High Priority | Good effort-to-impact ratio

Example 3: Enterprise SSO Integration

Problem:Reach: 500 (enterprise accounts). Impact: 3 (Massive - required for deals). Confidence: 90%. Effort: 6 person-weeks.

Solution:RICE = (500 × 3 × 0.90) / 6 RICE = 1,350 / 6 RICE = 225 Low score despite massive impact because: - Small reach (only 500 accounts) - High effort (6 weeks) BUT: Strategic consideration - May unlock $500K ARR in blocked deals - Revenue per account high RICE undervalues revenue-driving features. Adjust with business impact multiplier or use separate enterprise prioritization.

Result:RICE: 225 | Low Score BUT strategic | Consider revenue impact separately

Frequently Asked Questions

What is the RICE scoring model?

RICE (Reach, Impact, Confidence, Effort) is a prioritization framework developed by Intercom. Score = (Reach × Impact × Confidence%) / Effort. Higher scores indicate features worth building first.

What's a good RICE score?

Depends on your scale. Compare relative scores within your feature list. Scores vary widely by Reach (company size matters). Focus on ranking rather than absolute numbers.

How does RICE compare to other frameworks?

RICE is more quantitative than MoSCoW or Kano. ICE (similar but no Reach) is simpler. WSJF (Weighted Shortest Job First) is SAFe's version. RICE suits product teams with usage data.

What are RICE limitations?

Limitations: subjective inputs, doesn't capture dependencies, ignores strategic fit, and can favor incremental over transformational. Use alongside qualitative judgment and strategy alignment.

Should I use RICE for all decisions?

Best for: feature prioritization, roadmap planning, comparing similar-sized initiatives. Less suited for: bug fixes (just fix them), technical debt, or binary decisions (build vs not).

How often should I recalculate RICE?

Recalculate when: new data available, scope changes, confidence increases (after research), or quarterly planning. Update Effort if implementation reveals complexity.

Background & Theory

RICE quantifies feature priority using four factors: Reach (how many users affected), Impact (magnitude of effect), Confidence (certainty of estimates), and Effort (cost to build). The formula produces comparable scores across different feature types. Confidence acts as a risk adjustment, penalizing uncertain estimates.

History

RICE was developed by Sean McBride at Intercom in 2016 to address limitations of simpler scoring methods. It evolved from ICE (Impact, Confidence, Ease) by adding Reach to account for audience size. The framework gained widespread adoption in product management, particularly at data-driven SaaS companies.