Product-org-os prioritize-features

Document Intelligence

This skill supports three modes: Create, Update, and Find.

Mode Detection

@path/to/prioritization.md

Threshold: >=85% auto-proceed | 70-84% state assumption | <70% ask user

Mode Behaviors

CREATE: Gather feature list, select framework(s), score, produce ranked output.

UPDATE:

1. Read existing prioritization (search if path not provided)
2. Preserve scores for unchanged items
3. Re-score modified or added items
4. Show diff summary: "Added: [items]. Re-scored: [items]. Unchanged: [items]."

FIND:

1. Search paths below for prioritization documents
2. Present results: title, framework used, item count, path
3. Ask: "Update one of these, or create new?"

Search Locations

• product/

• roadmap/

• planning/

• prioritization/

Gotchas

• Prioritization criteria must be stated explicitly — different frameworks (RICE, ICE, Kano, MoSCoW, WSJF, Value vs Effort) give different results
• Never fabricate reach, impact, or confidence scores — use data or explicitly label as team estimates
• Prioritization without strategic alignment is just sorting — connect to strategic bets

Vision to Value Phase

Phase 3: Strategic Commitments - Prioritization converts decisions into executable commitments by ranking what to build and in what order.

Prerequisites: Phase 2 complete (strategic decisions made, business viability confirmed) Outputs used by:

/product-roadmap

/prd

/commitment-check

Methodology

Framework Selection Guide

When unsure, ask the user which framework to apply. If they say "just prioritize", default to RICE.

RICE Scoring

Formula: RICE Score = (Reach x Impact x Confidence) / Effort

Kano Classification

For each feature, ask the functional/dysfunctional question pair:

• Functional: "How would you feel if this feature were present?"
• Dysfunctional: "How would you feel if this feature were absent?"

Classification matrix (Functional x Dysfunctional):

M = Must-Be, O = One-Dimensional, A = Attractive, I = Indifferent, R = Reverse, Q = Questionable

Priority order: Must-Be > One-Dimensional > Attractive > Indifferent

MoSCoW Classification

WSJF Scoring

Formula: WSJF = (Business Value + Time Criticality + Risk Reduction) / Job Size

ICE Scoring

Formula: ICE Score = Impact × Confidence × Ease

Score interpretation: Max = 1000, practical range 1-500. Higher = prioritize first.

When to use over RICE: ICE is faster to score (no Reach estimation needed) and works well for growth experiments, A/B tests, and quick-iteration contexts where all items have similar reach. Use RICE when reach varies significantly across features.

Value vs Effort Matrix

Plot each feature on a 2×2 matrix with Value (vertical axis) and Effort (horizontal axis):

Scoring approach: Rate Value and Effort each on a 1-5 scale, then classify:

• Value >= 4, Effort <= 2 → Quick Win
• Value >= 4, Effort >= 3 → Big Bet
• Value <= 3, Effort <= 2 → Fill-In
• Value <= 3, Effort >= 3 → Money Pit

Priority order: Quick Wins > Big Bets > Fill-Ins > Money Pit (avoid)

When to use: Best for initial triage with stakeholders, executive-level alignment, or when you need a fast visual prioritization without detailed scoring. Often used as a first pass before applying RICE or WSJF to the Quick Wins and Big Bets quadrants.

Output Structure

# Feature Prioritization: [Context/Product Name]

**Date**: [YYYY-MM-DD]
**Owner**: [Who owns this prioritization]
**Framework(s)**: [RICE / ICE / Kano / MoSCoW / WSJF / Value vs Effort]
**Input source**: [Backlog, stakeholder request, etc.]

## Features Under Consideration

| # | Feature/Initiative | Description |
|---|-------------------|-------------|
| 1 | [Feature name] | [Brief description] |
| 2 | [Feature name] | [Brief description] |

## Scoring

### [Framework Name] Scores

[Framework-specific scoring table — see framework sections above]

## Ranked Results

| Rank | Feature | Score | Category/Tier | Rationale |
|------|---------|-------|---------------|-----------|
| 1 | [Feature] | [Score] | [Must/High/etc.] | [Why it ranked here] |
| 2 | [Feature] | [Score] | [Should/Med/etc.] | [Why it ranked here] |

## Key Insights

- **Top priority**: [Feature] because [reason]
- **Surprising result**: [Feature] ranked [higher/lower] than expected because [reason]
- **Tension**: [Feature A] vs [Feature B] — [tradeoff description]

## Assumptions & Caveats

- [Key assumptions that affect the scoring]
- [Data gaps that reduce confidence]
- [Recommendations for improving confidence]

## Next Steps

- [ ] Validate scores with [stakeholders]
- [ ] Feed top priorities into roadmap via `/product-roadmap`
- [ ] Design experiments for low-confidence items via `/experiment-design`

Instructions

1. Ask the user for: (a) the list of features/initiatives, (b) which framework(s) to use (default: RICE)
2. If the user provides features without descriptions, ask for brief descriptions
3. For RICE: ask the user to estimate Reach and Effort; propose Impact and Confidence based on context
4. For ICE: score all three dimensions 1-10; ideal for growth experiments and rapid prioritization
5. For Kano: note that ideal Kano requires customer survey data; offer to classify based on product knowledge as a proxy
6. For MoSCoW: facilitate classification discussion; challenge "Must Have" inflation
7. For WSJF: use relative sizing (Fibonacci); anchor with the smallest item as 1
8. For Value vs Effort: rate each dimension 1-5; classify into quadrants; best as a first-pass triage
9. Multiple frameworks can be applied to the same list for cross-validation
10. Save output as markdown file
11. Offer

    /product-roadmap

    to convert prioritized list into a roadmap

Integration

• Links to

  /product-roadmap

  (prioritized features feed roadmap themes)
• Links to

  /commitment-check

  (validate that prioritized commitments are achievable)
• Links to

  /assumption-map

  (surface assumptions behind scoring)
• Links to

  /experiment-design

  (test assumptions for low-confidence scores)
• Links to

  /context-recall

  (check past prioritization decisions for consistency)