Feature Prioritization (RICE) AI Prompts for PMs

TL;DR

AI prompts help product managers pressure-test RICE scores before making prioritization decisions
RICE framework components (Reach, Impact, Confidence, Effort) each require careful estimation
Scenario analysis reveals how sensitive prioritization is to estimation uncertainty
Stakeholder alignment is easier when prioritization decisions are grounded in explicit assumptions
Regular score recalibration keeps prioritization current with changing conditions

Introduction

Product managers spend an inordinate amount of time in prioritization limbo—backlogs growing faster than they can clear, stakeholders pulling in different directions, and the nagging sense that the most important work might not actually be getting done. The RICE framework offers a structured approach: quantifies Reach, Impact, Confidence, and Effort to generate comparable scores. But here’s the problem most PMs discover quickly: RICE scores are only as good as the estimates feeding them, and those estimates are often guesses dressed up in quantitative clothing.

The real value of RICE is not the final number—it’s the discipline of making assumptions explicit and pressure-testing them. When you force yourself to estimate Reach in users per quarter, Impact on a 0-4 scale, Confidence as a percentage, and Effort in person-months, you surface disagreements and uncertainties that would otherwise hide in qualitative debate. AI can accelerate this process by generating prompts that help you think through each component systematically, challenge your assumptions, and explore scenarios that reveal how robust your prioritization truly is.

This guide provides AI prompts designed specifically for product managers who want to move beyond rote RICE calculation to genuinely rigorous prioritization. Use these prompts to stress-test your scores, align stakeholders, and make prioritization decisions with greater confidence.

RICE Framework Fundamentals
Reach Estimation
Impact Assessment
Confidence Calibration
Effort Estimation
Scenario and Sensitivity Analysis
Stakeholder Alignment
Score Recalibration
FAQ: RICE Prioritization Excellence

RICE Framework Fundamentals {#rice-fundamentals}

Before diving into component estimation, establish clear principles for RICE calculation.

Prompt for RICE Methodology Assessment:

Assess our current RICE methodology:

CURRENT PRACTICE:
- How we estimate Reach
- How we assess Impact
- How we calibrate Confidence
- How we measure Effort

Framework assessment:

1. CONSISTENCY:
   - Are all teams using the same scale definitions?
   - Do PMs interpret scales the same way?
   - Are estimates based on actual data or gut feel?

2. COMPARABILITY:
   - Can scores from different PMs be compared?
   - Are feature types (technical debt, growth, UX) scored differently?
   - Do you apply any normalization across categories?

3. VALIDATION:
   - Do historical scores predict actual outcomes?
   - How often do you revisit and update scores?
   - What happens when scores conflict with stakeholder opinions?

4. INTEGRATION:
   - Is RICE the primary decision framework or one input among many?
   - How do you handle features that score low but stakeholders strongly want?
   - Is the framework helping or just providing false precision?

Provide specific recommendations for improving your RICE methodology.

Prompt for RICE Score Review:

Review the following RICE scores for a feature initiative:

FEATURE: [FEATURE DESCRIPTION]
CURRENT SCORES:
- Reach: [SCORE]
- Impact: [SCORE]
- Confidence: [PERCENTAGE]
- Effort: [PERSON-MONTHS]
- RICE Score: [CALCULATED]

Review dimensions:

1. REACH REALISM:
   - Is this reach plausible given user base and adoption?
   - What assumptions is this reach based on?
   - How sensitive is reach to different scenarios?

2. IMPACT CALIBRATION:
   - Is impact appropriately scaled (0.25, 0.5, 1, 2, 3, 4)?
   - What would immediate impact versus long-term impact look like?
   - How does this compare to similar past features?

3. CONFIDENCE JUSTIFICATION:
   - What is driving the confidence percentage?
   - Is low confidence a red flag for proceeding?
   - What information would increase confidence?

4. EFFORT REALISM:
   - How confident are we in the effort estimate?
   - What is the best/worst/expected case?
   - Are we accounting for all team members' time?

Provide specific challenges to each score and recommendations for recalibration.

Reach Estimation {#reach-estimation}

Reach is often the most uncertain component and the one most prone to optimistic bias.

Prompt for Reach Estimation:

Estimate Reach for the following feature:

FEATURE: [FEATURE DESCRIPTION]
USER BASE: [CURRENT USERS/ACTIVE USERS/RELEVANT SEGMENT]
LAUNCH TIMING: [QUARTER OF LAUNCH]

Reach estimation framework:

1. BASELINE CALCULATION:
   - Current active users in target segment
   - Expected initial adoption rate
   - Time to reach half of potential users

2. ADOPTION FUNNEL:
   - Awareness to activation rate
   - Activation to regular use rate
   - Churn and re-adoption dynamics

3. SEGMENT FILTERING:
   - What percentage of users will this feature apply to?
   - Is the segment growing or shrinking?
   - Are there geographic, demographic, or behavioral filters?

4. TIMING FACTORS:
   - Seasonal variations in usage
   - Growth trajectory between now and launch
   - Competitive dynamics that might affect adoption

Generate reach estimates for optimistic, expected, and pessimistic scenarios with specific assumptions for each.

Prompt for Reach Sensitivity Analysis:

Analyze reach sensitivity for the following feature:

FEATURE: [FEATURE DESCRIPTION]
BASE REACH ESTIMATE: [YOUR ESTIMATE]

Sensitivity factors:

1. ADOPTION RATE VARIANCE:
   - What if adoption is half of estimated?
   - What if adoption exceeds expectations?
   - At what point does reach justify the investment?

2. USER SEGMENT VARIANCE:
   - What if the applicable segment is smaller?
   - What if the segment grows faster/slower?
   - How does mobile versus desktop usage affect reach?

3. COMPETITIVE DISPLACEMENT:
   - What if a competitor releases something similar first?
   - What if competitive response reduces adoption?
   - What is our competitive advantage in driving reach?

4. INTERNAL DEPENDENCIES:
   - What if required integrations delay launch?
   - What if marketing support is less than planned?
   - How do internal factors affect reach confidence?

Identify which factors most affect reach and how robust the estimate is to variance.

Impact Assessment {#impact-assessment}

Impact estimation is inherently subjective but can be structured more rigorously.

Prompt for Impact Framework Development:

Develop clearer Impact assessment guidelines for your team:

IMPACT SCALE:
- 0.25: Minimal impact
- 0.5: Low impact
- 1: Medium impact (baseline)
- 2: High impact
- 3: Very high impact
- 4: Massive impact

Impact dimensions:

1. USER VALUE:
   - Task completion improvement
   - Time saved per user
   - Error reduction
   - User satisfaction improvement

2. BUSINESS METRICS:
   - Conversion rate impact
   - Retention improvement
   - Revenue per user
   - Cost reduction

3. STRATEGIC VALUE:
   - Competitive positioning
   - Market expansion
   - Brand perception
   - Platform ecosystem health

For each level, provide:
- Specific examples from past features
- Metrics that would validate the impact level
- Common mistakes in assessing this dimension

Create rubrics that help PMs assign consistent Impact scores.

Prompt for Impact Attribution:

Analyze impact attribution for the following feature:

FEATURE: [FEATURE DESCRIPTION]
EXPECTED IMPACT: [YOUR IMPACT ESTIMATE]

Attribution challenges:

1. MULTI-TOUCH ATTRIBUTION:
   - How much impact belongs to this feature versus others?
   - How do you isolate feature impact from other changes?
   - What happens if features ship simultaneously?

2. BASELINE COMPARISON:
   - What is the counterfactual (what would happen without this feature)?
   - How do you measure lift when you cannot A/B test?
   - What historical data supports the impact estimate?

3. LEAD VERSUS LAG INDICATORS:
   - Short-term versus long-term impact
   - Direct versus indirect impact
   - User-visible versus internal impact

4. IMPACT DECAY:
   - Does impact persist over time or diminish?
   - What is the impact trajectory over months/quarters?
   - How does impact vary by user segment?

Provide recommendations for more rigorous impact estimation.

Confidence Calibration {#confidence-calibration}

Confidence scores reveal how certain you are about your estimates.

Prompt for Confidence Calibration:

Evaluate the Confidence score for the following feature:

FEATURE: [FEATURE DESCRIPTION]
CURRENT CONFIDENCE: [PERCENTAGE]

Confidence factors:

1. DATA QUALITY:
   - Are reach/impact estimates based on actual data or guesses?
   - How much historical precedent supports these estimates?
   - Is the data from similar contexts or a stretch?

2. ESTIMATOR EXPERIENCE:
   - How much experience do we have with this type of feature?
   - Are we estimating within our core competency or venturing into new territory?
   - How confident are other team members in this estimate?

3. ASSUMPTION DEPENDENCY:
   - How many assumptions are embedded in the estimate?
   - How uncertain are each of those assumptions?
   - Would different assumptions dramatically change the score?

4. ENVIRONMENT STABILITY:
   - Is this a new context or familiar territory?
   - How stable are the factors this estimate depends on?
   - What external changes could invalidate the estimate?

Provide a more calibrated confidence score with specific justification.

Prompt for Low-Confidence Decision Framework:

Develop guidance for features with low Confidence scores:

LOW-CONFIDENCE SCENARIO:
- Feature: [DESCRIPTION]
- Confidence: [PERCENTAGE]
- RICE Score: [SCORE]

Decision framework:

1. WHEN TO PROCEED:
   - Can the feature be structured to reduce uncertainty?
   - Is this exploratory work that is inherently uncertain?
   - Is the potential impact high enough to justify uncertainty?

2. WHEN TO DELAY:
   - What information would most increase confidence?
   - Is there a way to get that information before committing?
   - What is the cost of waiting versus the cost of uncertainty?

3. RISK MITIGATION:
   - Can we structure this as an experiment?
   - Can we build incrementally to reduce risk?
   - What monitoring would catch problems early?

4. TRANSPARENT COMMUNICATION:
   - How do we communicate uncertainty to stakeholders?
   - What guardrails do we put in place?
   - How do we set expectations appropriately?

Provide a recommendation for handling this low-confidence feature.

Effort Estimation {#effort-estimation}

Effort is often underestimated, especially for complex features.

Prompt for Comprehensive Effort Estimation:

Develop a rigorous effort estimate for:

FEATURE: [FEATURE DESCRIPTION]
TEAM: [TEAM COMPOSITION AND CAPACITY]

Effort components:

1. CORE DEVELOPMENT:
   - Feature development
   - Testing and QA
   - Documentation

2. INDIRECT WORK:
   - Design and UX
   - Technical architecture
   - Code review and merge
   - Deployment and launch

3. CONTINGENCY:
   - Bug fixes and edge cases
   - Integration complications
   - Performance optimization
   - Accessibility and internationalization

4. OVERHEAD:
   - Meetings and coordination
   - Knowledge transfer
   - Onboarding new team members
   - Administrative tasks

Generate effort estimate in person-months with best case, expected case, and worst case scenarios.

Prompt for Technical Debt in Effort:

Analyze how technical debt should factor into effort estimates:

FEATURE: [FEATURE DESCRIPTION]
STATED EFFORT: [ESTIMATE]

Technical debt considerations:

1. EXISTING DEBT:
   - How much existing debt does this feature touch?
   - Will we need to pay down debt to implement this feature?
   - How does debt repayment time compare to feature time?

2. NEW DEBT CREATION:
   - Will this feature create new technical debt?
   - Is the debt acceptable for the value delivered?
   - What is the plan for addressing new debt?

3. DEBT VERSUS FEATURE TRADE-OFFS:
   - Should we reduce feature scope to pay down debt?
   - Can we defer debt payment to a future sprint?
   - What is the carrying cost of the debt?

Recommend how to handle technical debt in your effort estimate.

Scenario and Sensitivity Analysis {#scenario-analysis}

RICE scores are point estimates; sensitivity analysis reveals how robust decisions are.

Prompt for Sensitivity Analysis:

Conduct sensitivity analysis on the following RICE scores:

FEATURES AND SCORES:
[LIST FEATURES WITH REACH, IMPACT, CONFIDENCE, EFFORT, RICE]

Analysis:

1. BEST/WORST CASE ANALYSIS:
   - What if all estimates are optimistic?
   - What if all estimates are pessimistic?
   - How does relative ranking change?

2. VARIABLE SENSITIVITY:
   - Which variable (Reach, Impact, Confidence, Effort) most affects rankings?
   - How much would that variable need to change to change the decision?
   - Which estimates should we validate most carefully?

3. THRESHOLD ANALYSIS:
   - What score threshold separates "do now" from "do later"?
   - How much buffer exists around that threshold?
   - What would need to happen for borderline features to cross the threshold?

4. MONTE CARLO SIMULATION:
   - If you ran 1000 simulations with varying estimates, what percentage would change the decision?
   - How confident can you be in the prioritization?

Identify which prioritization decisions are robust versus fragile.

Prompt for Scenario Planning:

Develop scenario-based prioritization for:

INITIATIVE: [DESCRIBE THE INITIATIVE]

Scenarios:

1. OPTIMISTIC CASE:
   - Reach exceeds expectations
   - Impact is high
   - Effort is low
   - Confidence is high
   - What wins? What else becomes viable?

2. PESSIMISTIC CASE:
   - Reach underperforms
   - Impact is low
   - Effort overruns
   - Confidence was misplaced
   - What still makes sense to build?

3. MOST LIKELY CASE:
   - Reasonable estimates across all dimensions
   - What should we prioritize given this balanced view?

4. SURPRISE SCENARIO:
   - Something unexpected happens (competitor, market, technology)
   - How does our prioritization hold up?
   - What would we wish we had built?

For each scenario, provide RICE-inspired analysis and strategic implications.

Stakeholder Alignment {#stakeholder-alignment}

RICE helps align stakeholders when used as a discussion framework, not just a score.

Prompt for Stakeholder Score Reconciliation:

Facilitate stakeholder alignment on the following feature:

FEATURE: [FEATURE DESCRIPTION]
DIFFERING SCORES:
- PM Score: [SCORE]
- Engineering estimate: [SCORE]
- Sales input: [SCORE]
- Customer input: [SCORE]

Reconciliation framework:

1. ASSUMPTION EXPLICITNESS:
   - What assumptions is each stakeholder making?
   - Where do assumptions differ?
   - Which assumptions are most critical?

2. PERSPECTIVE LEGITIMACY:
   - Does each perspective reflect legitimate concerns?
   - Are some perspectives overweighted?
   - How do we weight different stakeholder inputs?

3. INFORMATION GAPS:
   - What information would resolve the disagreement?
   - Can we get that information quickly?
   - Should we proceed with uncertainty or wait?

4. DECISION FRAMEWORK:
   - Is RICE the right framework for this decision?
   - What other factors should influence the decision?
   - How do we make a decision when stakeholders disagree?

Develop a path to alignment or a clear decision framework.

Prompt for RICE Communication:

Develop a RICE communication approach for stakeholders:

AUDIENCE: [WHO YOU ARE PRESENTING TO]

Communication objectives:
1. Help them understand the RICE framework
2. Share your prioritization recommendations
3. Address likely objections
4. Get alignment on next steps

Presentation structure:

1. FRAMEWORK INTRODUCTION (if needed):
   - What RICE is and why we use it
   - How scores are estimated
   - What the scores do and do not capture

2. FEATURE REVIEW:
   - Top features by RICE score
   - Key assumptions driving scores
   - Trade-offs being made

3. UNCERTAINTY DISCUSSIONS:
   - Where confidence is low
   - What could change scores
   - Monitoring we will implement

4. RECOMMENDATION:
   - Clear prioritization recommendation
   - What we are not doing and why
   - What we need from stakeholders

Develop stakeholder-specific communication materials.

Score Recalibration {#score-recalibration}

RICE scores become stale; recalibration keeps prioritization current.

Prompt for Score Recalibration Triggers:

Develop triggers for when to recalculate RICE scores:

SCORE TRIGGERS:

1. TIME-BASED:
   - When should scores automatically be revisited?
   - How stale is too stale?
   - Quarterly? Monthly? Per-sprint?

2. INFORMATION-BASED:
   - What new information should trigger recalculation?
   - Customer research findings
   - Competitive developments
   - Technical discoveries

3. VARIANCE-BASED:
   - When actual results deviate from estimates
   - What tracking reveals about estimation accuracy
   - How to adjust future estimates based on variance

4. CONTEXT-BASED:
   - When strategic priorities shift
   - When resources change
   - When timeline expectations change

Create a recalibration framework that keeps scores current without constant rework.

Prompt for Estimation Accuracy Analysis:

Analyze the accuracy of your historical RICE estimates:

PROCESS:
1. Pull past features with RICE scores
2. Compare estimated scores to actual outcomes
3. Analyze where estimates were wrong
4. Adjust estimation processes

Analysis dimensions:

1. REACH ACCURACY:
   - Were actual users higher or lower than estimated?
   - By what percentage did estimates miss?
   - What explains the variance?

2. IMPACT ACCURACY:
   - Did features achieve expected impact?
   - What impact metrics were measured?
   - How should future impact estimates be calibrated?

3. EFFORT ACCURACY:
   - Did features take more or less effort than estimated?
   - What was the average variance?
   - Are there patterns in overruns/underruns?

4. OVERALL RICE ACCURACY:
   - Did features with higher scores perform better?
   - Was prioritization decision-making improved?
   - What changes to the framework are indicated?

Provide specific recommendations for improving estimation accuracy.

FAQ: RICE Prioritization Excellence {#faq}

Should RICE replace qualitative stakeholder judgment?

No. RICE should inform and structure judgment, not replace it. The framework makes assumptions explicit and enables consistent comparison, but the final decision should incorporate factors RICE cannot capture: strategic priorities, organizational politics, dependencies, and circumstances specific to the moment. Use RICE as a powerful input to decision-making, not the decision itself.

How do I handle features that score low but stakeholders strongly want?

First, understand why stakeholders want it despite the low score. Often this reveals information the score does not capture—strategic importance, customer commitments, or political considerations. Discuss the discrepancy openly using RICE language. If stakeholders still want it, ensure they understand the trade-offs being made. Sometimes legitimate factors outside RICE should override the score.

What is a reasonable confidence level for proceeding?

Higher confidence (80-100%) is appropriate when estimates are based on solid data and historical precedent. Lower confidence (50-70%) may be acceptable for exploratory work, innovative features, or situations where you are deliberately accepting uncertainty. Below 50% confidence should prompt serious consideration of whether you have enough information to prioritize confidently.

How often should I recalculate RICE scores?

Reassess scores monthly for the current quarter’s features. Quarterly features should have scores refreshed quarterly. Strategic features with long timelines should be reviewed when significant new information emerges. The key is distinguishing between features that are actively being worked (requiring current scores) versus those in the backlog (requiring periodic refresh).

Should I normalize RICE scores across different feature types?

Yes, if different types of features (growth, platform, UX, technical debt) are competing for the same resources. Without normalization, technical debt often scores poorly on Reach and Impact despite being essential. Consider separate scoring rubrics or adjustment factors for different feature types that reflect their different value patterns.

Conclusion

RICE is a powerful prioritization framework when used rigorously—not as a score generator, but as a discipline for making assumptions explicit and pressure-testing them. The prompts in this guide help you move beyond rote calculation to genuine analytical rigor.

Key Takeaways:

RICE scores are only as good as the estimates feeding them—invest in estimation quality.
Confidence scores reveal uncertainty worth acknowledging—low confidence should change behavior, not just be noted.
Sensitivity analysis reveals how robust decisions are—identify which decisions could flip with better information.
Stakeholder alignment is easier with explicit assumptions—RICE enables structured discussion.
Recalibration keeps prioritization current—stale scores lead to stale decisions.

Next Steps:

Audit your current RICE estimation practices against this guide
Develop clearer rubrics for each RICE component
Implement sensitivity analysis for key prioritization decisions
Create stakeholder communication materials that explain your approach
Establish recalibration triggers for your backlog

RICE done right brings discipline to prioritization. RICE done wrong provides false precision that undermines trust. Use these prompts to ensure your RICE practice earns the trust it needs to drive good decisions.

Feature Prioritization (RICE) AI Prompts for PMs