SaaS Feature Prioritization Framework - RICE Scoring Worksheet

R: Reach

Definition: How many people will this initiative affect within a specific time period?

How to score:

• Pick a timeframe (typically 1 quarter or 1 month)
• Estimate the number of users, customers, or events the feature will impact in that period
• Use real numbers from your analytics (not guesses)

Examples of reach:

• "500 customers per quarter" — if the feature appears in a flow 500 customers complete monthly
• "150 new signups per month" — if the feature is part of onboarding and 150 people sign up monthly
• "2,000 page views per month" — if the feature affects a page with that traffic

Rules:

• Use the same timeframe for all features you're comparing
• Be honest — don't count your entire user base if only 10% will ever see the feature
• Distinguish between one-time reach (launch) and recurring reach (ongoing)

Common mistake: Assuming 100% of users will encounter a new feature. Most features affect only a subset.

I: Impact

Definition: How much will this initiative impact each person who encounters it?

Intercom's official 5-tier scoring scale (use this):

How to score:

• Consider the magnitude of change the feature creates for each affected user
• Think about both immediate impact and long-term behavior change
• Use quantitative data when possible (e.g., "will likely reduce time-to-complete by 40%")

Examples:

• Massive (3): Adding a feature that unblocks a common frustration and enables a workflow that wasn't possible before
• High (2): Making a complex process 2x faster
• Medium (1): A meaningful but incremental improvement to an existing workflow
• Low (0.5): Polish improvements, minor UX fixes
• Minimal (0.25): Tiny visual tweaks, edge case fixes

Common mistake: Defaulting everything to "High" impact. Be disciplined — most features are "Medium" at best.

C: Confidence

Definition: How sure are you about your Reach, Impact, and Effort estimates?

Intercom's official 3-tier scoring scale (use this):

Anything below 50% Intercom calls a "moonshot" — too speculative to prioritize via RICE.

How to score:

• Ask: "If I presented this score to a skeptical executive, could I defend my assumptions with data?"
• Be honest about what's backed by research vs. what's a guess
• Don't inflate confidence to make a feature look better

Confidence anchors:

• 100% High: You have A/B test data, analytics from a similar feature, or direct user research validating both reach and impact
• 80% Medium: You have solid data for 2 of the 3 estimates (R, I, E) and reasonable assumptions for the third
• 50% Low: You have assumptions for most estimates or you're exploring a completely new area

Common mistake: Treating confidence as optional. Low confidence should significantly reduce a feature's priority — that's the whole point of this factor.

E: Effort

Definition: How many "person-months" will this take, totaled across design, engineering, testing, and product?

How to score:

• Estimate person-months (PM) across all roles
• Use whole numbers for anything over a month
• Use 0.5 for anything under a month
• Don't get into the weeds of precise estimates — ballpark is fine

Examples:

• 0.5 PM: A 1-2 week tweak or small feature
• 1 PM: ~1 month of total team effort
• 2 PM: ~2 months of work split across team
• 5 PM: A major initiative requiring full team for over a month

Who contributes to effort? Usually:

• Product manager (for spec, requirements, QA)
• Designer (for UX, UI, prototyping)
• Engineer(s) (frontend, backend)
• QA / tester
• Other specialists (data, infra) if applicable

Common mistake: Underestimating effort. Always ask engineers for their estimate — don't guess.

A product team is deciding between 5 feature ideas. Here's how RICE helps prioritize:

Feature A: Improve Email Verification Flow

• Reach: 1,200/mo (new signups)
• Impact: 2 (High — reduces early churn)
• Confidence: 100% (data from signup funnel analysis)
• Effort: 1 PM
• RICE: (1200 × 2 × 1.0) / 1 = 2,400

Feature B: Build AI-Powered Report Generator

• Reach: 200/mo (power users who generate reports)
• Impact: 3 (Massive — saves hours per week)
• Confidence: 50% (speculative — unclear if users want AI here)
• Effort: 5 PM
• RICE: (200 × 3 × 0.50) / 5 = 60

Feature C: Add Dark Mode

• Reach: 2,500/mo (all active users eventually)
• Impact: 0.5 (Low — polish, not core value)
• Confidence: 80% (user requests are frequent)
• Effort: 1 PM
• RICE: (2500 × 0.5 × 0.80) / 1 = 1,000

Feature D: Redesign Dashboard

• Reach: 2,500/mo (all active users)
• Impact: 1 (Medium — improved but not game-changing)
• Confidence: 80% (user research supports direction)
• Effort: 4 PM
• RICE: (2500 × 1 × 0.80) / 4 = 500

Feature E: Add Slack Integration

• Reach: 800/mo (estimated users who'd use Slack integration)
• Impact: 2 (High — unblocks workflow)
• Confidence: 100% (top-requested feature in surveys)
• Effort: 2 PM
• RICE: (800 × 2 × 1.0) / 2 = 800

Priority Order (by RICE score):

1. Feature A: Email Verification Flow — 2,400
2. Feature C: Dark Mode — 1,000
3. Feature E: Slack Integration — 800
4. Feature D: Dashboard Redesign — 500
5. Feature B: AI Report Generator — 60

Insights from this example:

• Feature A wins easily because it affects all new users with high confidence and low effort
• Feature B (the "exciting" AI idea) scores low because of high effort + low confidence
• Feature C (Dark Mode) scores surprisingly high — easy to build, high reach — but delivers low impact
• This comparison forces the team to recognize that Dark Mode's "high priority" from user requests might not match business impact

This is the value of RICE: It surfaces trade-offs that intuition misses.

RICE works well for:

• ✅ Comparing similar types of features
• ✅ Building a quarterly roadmap
• ✅ Removing bias from prioritization debates
• ✅ Creating a shared team language for "what's important"
• ✅ Small-to-medium initiatives (1-10 PM of effort)

RICE doesn't work well for:

• ❌ Strategic bets / moonshot initiatives (confidence always too low)
• ❌ Platform investments (reach is hard to quantify — benefits are indirect)
• ❌ Technical debt (impact isn't directly user-facing)
• ❌ Legal/compliance work (not optional regardless of score)
• ❌ Very large initiatives that span quarters
• ❌ Creative/brand initiatives (hard to quantify impact)

For things RICE doesn't capture well:

• Strategic initiatives: Use a separate "strategic roadmap" with its own prioritization
• Tech debt: Allocate a fixed percentage of capacity (e.g., 20% of every sprint)
• Compliance/legal: Non-negotiable — do these regardless of RICE
• Moonshots: Reserve ~10% of capacity for high-risk, high-reward bets

Run this workshop quarterly to prioritize your backlog:

Participants: PM, Design lead, Engineering lead, optional CEO/founder Time: 2 hours Prep: List of 10-20 candidate features with brief descriptions

Agenda:

0:00–0:15 — Review Methodology

• Recap what Reach, Impact, Confidence, Effort mean
• Align on the timeframe for reach (e.g., "per quarter")
• Agree on impact scale interpretations

0:15–1:00 — Score Each Feature

• For each feature, discuss Reach, Impact, Confidence, Effort
• Use data when available
• When team disagrees, discuss — don't average
• Document assumptions ("We assumed 500 users because X")

1:00–1:20 — Calculate & Rank

• Compute RICE scores
• Rank features high-to-low
• Review the ranking — does it match intuition?

1:20–1:40 — Sanity Check

• Anything unexpectedly high? (Might be over-scored)
• Anything unexpectedly low? (Might have unmeasured strategic value)
• Adjust if clearly wrong — but require justification

1:40–2:00 — Commit to Top N

• Based on quarterly capacity, commit to top features
• Remaining features stay in backlog for next quarter
• Assign owners and rough timelines

• Sean McBride, "RICE: Simple Prioritization for Product Managers," Intercom (original article, 2016)
• Intercom Product Team, "How We Made Our Prioritization Framework"
• ProductPlan, "RICE Scoring Model" documentation
• Dan Olsen, "The Lean Product Playbook"
• Marty Cagan, "Inspired: How to Create Tech Products Customers Love"

Created by Desisle — SaaS UI/UX Design Agency desisle.com | hello@desisle.com Free to use and share with attribution.

For a custom product prioritization workshop, contact us at hello@desisle.com.