RICE
Score your initiatives with Reach, Impact, Confidence and Effort. Intercom's prioritization model to arbitrate your roadmap without politics.
Description
RICE scoring is a quantitative prioritization model created by Sean McBride at Intercom in 2018, which evaluates each product initiative across four factors (Reach, Impact, Confidence, Effort) to produce a comparable score for objectively arbitrating a roadmap. Its purpose: to turn opinion-based debates into a defensible calculation when 30 initiatives are competing for the next 5 spots in the backlog. The formula is straightforward: RICE Score = (Reach x Impact x Confidence) / Effort. Reach measures how many users will be affected by the initiative over a given period (e.g., 500 users per quarter). Impact estimates the effect on each user, rated on a scale from 0.25 (minimal) to 3 (massive). Confidence reflects your degree of certainty about these estimates, expressed as a percentage (80% if you have data, 50% if it is a gut feeling). Effort quantifies the work required in person-months. McBride explained it in his foundational article on the Intercom blog: "We needed a scoring system that let us compare wildly different ideas in a consistent way." RICE scoring works like a precision scale in a laboratory: every initiative is weighed on the same four criteria, making the result challengeable on the data but not on the method. What sets RICE apart from other prioritization frameworks (ICE, MoSCoW, WSJF) is the Confidence factor. By forcing each team to make their level of certainty explicit, RICE penalizes attractive but unvalidated ideas and rewards initiatives backed by data. Intercom used it to narrow 100+ ideas down to a quarterly roadmap in a few hours. Atlassian and ProductPlan later integrated it as a native feature in their product management tools.
Objectives
- Prioritize features
- Prioritize work
- Ensure strategic alignment
Used by
- -Intercom (creator of the RICE model, used internally to prioritize its product roadmap since 2018)
- -Atlassian (integrates RICE scoring as a native feature in its product management tools)
Advantages
- Data-driven prioritization. The numerical score replaces opinion debates and allows comparing radically different initiatives on the same scale.
- The Confidence factor penalizes uncertainty. Unlike ICE or MoSCoW, RICE forces quantifying what you don't know, which protects the roadmap from unvalidated "pet projects".
- Easily communicable to stakeholders. A table sorted by RICE score is understandable by the CEO, CTO and salesperson in 2 minutes.
- Compatible with any tool. A spreadsheet is enough. No need to invest in a dedicated tool to get started.
Limitations
- Scoring quality directly tied to estimation quality. If Reach is invented and Confidence overestimated, the RICE score gives a false precision that is worse than no score at all.
- Does not capture dependencies between initiatives. Two initiatives may score high individually but be redundant, or a low-scoring initiative may be a technical prerequisite for three others.
- Can favor incremental optimizations. High Reach, low Effort initiatives (quick wins) naturally dominate the ranking, which can marginalize long-term strategic bets.
- Time-consuming on large backlogs. Scoring 100 initiatives across 4 dimensions takes time. Pre-filter with a quick MoSCoW before pulling out the RICE calculator.
How to apply RICE
- 1
List all candidate initiatives
Gather in a spreadsheet or dedicated tool all competing initiatives: features, improvements, major bugs, technical debt. Each row should be formulated as an expected outcome, not as a technical task. Aim for 15 to 50 initiatives. Beyond that, segment by theme. Output: unified list of initiatives to prioritize.
- 2
Estimate the Reach of each initiative
Quantify the number of users or customers who will be impacted by the initiative over a defined period (per quarter, per month). Use your analytics data: how many users touch this part of the product? How many customers have requested this feature? Be precise: "500 users/quarter" is a Reach, "a lot of people" is not. Output: quantified Reach per initiative.
- 3
Assess the Impact on each user
Rate the expected impact on a predefined scale: 3 = massive impact (transforms the experience), 2 = strong impact, 1 = moderate impact, 0.5 = low impact, 0.25 = minimal impact. If you hesitate between two levels, take the lower one. Impact measures the depth of change for each user affected, not the number of people concerned (that is the role of Reach). Output: Impact score per initiative.
- 4
Define the Confidence level
Express as a percentage your degree of certainty about the previous estimates. 100% = solid data (analytics, user studies). 80% = strong intuition backed by signals. 50% = reasonable hypothesis without data. Below 50%, consider that you need to validate the hypothesis before scoring. Confidence is the safeguard that prevents "pet projects" from monopolizing the roadmap. Output: Confidence percentage per initiative.
- 5
Estimate the Effort in person-months
Quantify the total work required: how many people, for how long? A project mobilizing 2 developers for 3 weeks = 1.5 person-months. Include design, QA, and documentation. Round to the nearest 0.5. An Effort of 0.5 = a few days for one person. An Effort of 10 = an entire quarter for a team. Output: Effort quantified in person-months per initiative.
- 6
Calculate the RICE score
Apply the formula: (Reach x Impact x Confidence%) / Effort. Example: an initiative reaching 1,000 users (Reach), with an Impact of 2, a Confidence of 80%, and an Effort of 3 person-months gives a score of (1000 x 2 x 0.8) / 3 = 533. Sort by descending score. Output: sorted table with RICE score per initiative.
- 7
Analyze the results and challenge the extremes
Review the top 10 and bottom 10. Very high scores deserve verification: is the Reach realistic? Is the Confidence justified? Very low scores deserve a second look: a long-term strategic initiative may have low Reach today but enormous Reach in 6 months. The RICE score is a decision input, not the decision itself. Output: top 10 validated and adjusted if necessary.
- 8
Decide and communicate the roadmap
Commit to the top 3 to 8 initiatives for the next cycle. Document the RICE score of each retained and non-retained initiative. Share the full table with stakeholders. When someone asks "why not my initiative?", the RICE score provides a factual answer. Revisit scores each quarter with new data. Output: prioritized roadmap communicated with justifications.