Files
gh-lyndonkl-claude/skills/decision-matrix/SKILL.md
2025-11-30 08:38:26 +08:00

183 lines
8.9 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
---
name: decision-matrix
description: Use when comparing multiple named alternatives across several criteria, need transparent trade-off analysis, making group decisions requiring alignment, choosing between vendors/tools/strategies, stakeholders need to see decision rationale, balancing competing priorities (cost vs quality vs speed), user mentions "which option should we choose", "compare alternatives", "evaluate vendors", "trade-offs", or when decision needs to be defensible and data-driven.
---
# Decision Matrix
## What Is It?
A decision matrix is a structured tool for comparing multiple alternatives against weighted criteria to make transparent, defensible choices. It forces explicit trade-off analysis by scoring each option on each criterion, making subjective factors visible and comparable.
**Quick example:**
| Option | Cost (30%) | Speed (25%) | Quality (45%) | Weighted Score |
|--------|-----------|------------|---------------|----------------|
| Option A | 8 (2.4) | 6 (1.5) | 9 (4.05) | **7.95** ← Winner |
| Option B | 6 (1.8) | 9 (2.25) | 7 (3.15) | 7.20 |
| Option C | 9 (2.7) | 4 (1.0) | 6 (2.7) | 6.40 |
The numbers in parentheses show criterion score × weight. Option A wins despite not being fastest or cheapest because quality matters most (45% weight).
## Workflow
Copy this checklist and track your progress:
```
Decision Matrix Progress:
- [ ] Step 1: Frame the decision and list alternatives
- [ ] Step 2: Identify and weight criteria
- [ ] Step 3: Score each alternative on each criterion
- [ ] Step 4: Calculate weighted scores and analyze results
- [ ] Step 5: Validate quality and deliver recommendation
```
**Step 1: Frame the decision and list alternatives**
Ask user for decision context (what are we choosing and why), list of alternatives (specific named options, not generic categories), constraints or dealbreakers (must-have requirements), and stakeholders (who needs to agree). Understanding must-haves helps filter options before scoring. See [Framing Questions](#framing-questions) for clarification prompts.
**Step 2: Identify and weight criteria**
Collaborate with user to identify criteria (what factors matter for this decision), determine weights (which criteria matter most, as percentages summing to 100%), and validate coverage (do criteria capture all important trade-offs). If user is unsure about weighting → Use [resources/template.md](resources/template.md) for weighting techniques. See [Criterion Types](#criterion-types) for common patterns.
**Step 3: Score each alternative on each criterion**
For each option, score on each criterion using consistent scale (typically 1-10 where 10 = best). Ask user for scores or research objective data (cost, speed metrics) where available. Document assumptions and data sources. For complex scoring → See [resources/methodology.md](resources/methodology.md) for calibration techniques.
**Step 4: Calculate weighted scores and analyze results**
Calculate weighted score for each option (sum of criterion score × weight). Rank options by total score. Identify close calls (options within 5% of each other). Check for sensitivity (would changing one weight flip the decision). See [Sensitivity Analysis](#sensitivity-analysis) for interpretation guidance.
**Step 5: Validate quality and deliver recommendation**
Self-assess using [resources/evaluators/rubric_decision_matrix.json](resources/evaluators/rubric_decision_matrix.json) (minimum score ≥ 3.5). Present decision-matrix.md file with clear recommendation, highlight key trade-offs revealed by analysis, note sensitivity to assumptions, and suggest next steps (gather more data on close calls, validate with stakeholders).
## Framing Questions
**To clarify the decision:**
- What specific decision are we making? (Choose X from Y alternatives)
- What happens if we don't decide or choose wrong?
- When do we need to decide by?
- Can we choose multiple options or only one?
**To identify alternatives:**
- What are all the named options we're considering?
- Are there other alternatives we're ruling out immediately? Why?
- What's the "do nothing" or status quo option?
**To surface must-haves:**
- Are there absolute dealbreakers? (Budget cap, timeline requirement, compliance need)
- Which constraints are flexible vs rigid?
## Criterion Types
Common categories for criteria (adapt to your decision):
**Financial Criteria:**
- Upfront cost, ongoing cost, ROI, payback period, budget impact
- Typical weight: 20-40% (higher for cost-sensitive decisions)
**Performance Criteria:**
- Speed, quality, reliability, scalability, capacity, throughput
- Typical weight: 30-50% (higher for technical decisions)
**Risk Criteria:**
- Implementation risk, reversibility, vendor lock-in, technical debt, compliance risk
- Typical weight: 10-25% (higher for enterprise/regulated environments)
**Strategic Criteria:**
- Alignment with goals, future flexibility, competitive advantage, market positioning
- Typical weight: 15-30% (higher for long-term decisions)
**Operational Criteria:**
- Ease of use, maintenance burden, training required, integration complexity
- Typical weight: 10-20% (higher for internal tools)
**Stakeholder Criteria:**
- Team preference, user satisfaction, executive alignment, customer impact
- Typical weight: 5-15% (higher for change management contexts)
## Weighting Approaches
**Method 1: Direct Allocation (simplest)**
Stakeholders assign percentages totaling 100%. Quick but can be arbitrary.
**Method 2: Pairwise Comparison (more rigorous)**
Compare each criterion pair: "Is cost more important than speed?" Build ranking, then assign weights.
**Method 3: Must-Have vs Nice-to-Have (filters first)**
Separate absolute requirements (pass/fail) from weighted criteria. Only evaluate options that pass must-haves.
**Method 4: Stakeholder Averaging (group decisions)**
Each stakeholder assigns weights independently, then average. Reveals divergence in priorities.
See [resources/methodology.md](resources/methodology.md) for detailed facilitation techniques.
## Sensitivity Analysis
After calculating scores, check robustness:
**1. Close calls:** Options within 5-10% of winner → Need more data or second opinion
**2. Dominant criteria:** One criterion driving entire decision → Is weight too high?
**3. Weight sensitivity:** Would swapping two criterion weights flip the winner? → Decision is fragile
**4. Score sensitivity:** Would adjusting one score by ±1 point flip the winner? → Decision is sensitive to that data point
**Red flags:**
- Winner changes with small weight adjustments → Need stakeholder alignment on priorities
- One option wins every criterion → Matrix is overkill, choice is obvious
- Scores are mostly guesses → Gather more data before deciding
## Common Patterns
**Technology Selection:**
- Criteria: Cost, performance, ecosystem maturity, team familiarity, vendor support
- Weight: Performance and maturity typically 50%+
**Vendor Evaluation:**
- Criteria: Price, features, integration, support, reputation, contract terms
- Weight: Features and integration typically 40-50%
**Strategic Choices:**
- Criteria: Market opportunity, resource requirements, risk, alignment, timing
- Weight: Market opportunity and alignment typically 50%+
**Hiring Decisions:**
- Criteria: Experience, culture fit, growth potential, compensation expectations, availability
- Weight: Experience and culture fit typically 50%+
**Feature Prioritization:**
- Criteria: User impact, effort, strategic value, risk, dependencies
- Weight: User impact and strategic value typically 50%+
## When NOT to Use This Skill
**Skip decision matrix if:**
- Only one viable option (no real alternatives to compare)
- Decision is binary yes/no with single criterion (use simpler analysis)
- Options differ on only one dimension (just compare that dimension)
- Decision is urgent and stakes are low (analysis overhead not worth it)
- Criteria are impossible to define objectively (purely emotional/aesthetic choice)
- You already know the answer (using matrix to justify pre-made decision is waste)
**Use instead:**
- Single criterion → Simple ranking or threshold check
- Binary decision → Pro/con list or expected value calculation
- Highly uncertain → Scenario planning or decision tree
- Purely subjective → Gut check or user preference vote
## Quick Reference
**Process:**
1. Frame decision → List alternatives
2. Identify criteria → Assign weights (sum to 100%)
3. Score each option on each criterion (1-10 scale)
4. Calculate weighted scores → Rank options
5. Check sensitivity → Deliver recommendation
**Resources:**
- [resources/template.md](resources/template.md) - Structured matrix format and weighting techniques
- [resources/methodology.md](resources/methodology.md) - Advanced techniques (group facilitation, calibration, sensitivity analysis)
- [resources/evaluators/rubric_decision_matrix.json](resources/evaluators/rubric_decision_matrix.json) - Quality checklist before delivering
**Deliverable:** `decision-matrix.md` file with table, rationale, and recommendation