size-opportunity

Skill: Size Opportunity

Purpose

Estimate the business impact and financial value of a given opportunity. Calls the opportunity-sizer agent to analyze addressable market, conversion potential, and revenue impact with clear assumptions and sensitivity analysis.

When to Use

• User asks "how much is this opportunity worth?"
• After identifying a potential improvement, quantify its value
• When evaluating which opportunities to prioritize
• When validating the ROI of a proposed project

Invocation

/size-opportunity {brief} — size an opportunity from description /size-opportunity --template {template} — use a sizing template (e.g., "product", "retention", "efficiency") /size-opportunity --sensitivity — run sensitivity analysis on key assumptions

Instructions

Step 1: Parse the Opportunity Brief

Extract:

1. What is the opportunity? — New feature, performance improvement, retention lever, new market, etc.
2. Who benefits? — User segment, customer tier, geography
3. Current state: — Baseline metric/revenue (query from data if available)
4. Proposed improvement: — What changes with this opportunity?
5. Timeline: — When could this launch? How long to maturity?

Step 2: Invoke Opportunity-Sizer Agent

Hand off to the opportunity-sizer agent with:

• The opportunity brief
• Current business metrics (active dataset context)
• User constraints and timeline
• Instructions to estimate impact and produce a sizing analysis

Step 3: Generate Sizing Analysis

The opportunity-sizer agent produces:

# Opportunity Sizing: {Opportunity Name}

## Executive Summary
**Opportunity:** {brief description}
**Addressable users:** {N} ({pct}% of base)
**Estimated annual impact:** ${X}M / {Y}% uplift
**Payback period:** {months}
**Confidence:** {High / Medium / Low}

---

## Detailed Breakdown

### Market / Addressable Opportunity

| Dimension | Current | Target | Addressable |
|-----------|---------|--------|-------------|
| Total users | {n} | — | — |
| Segment eligible | {n} | {pct}% of total | {n} |
| Will engage | {n} | {pct}% of segment | {n} |
| **Total addressable** | — | — | {n} users |

### Impact Model

**Revenue opportunity:**

{Addressable users} × {adoption rate}% × {ARPU improvement per user} = ${impact}

**Time to value:**
- **Build time:** {weeks}
- **Ramp time:** {weeks} to full adoption
- **Measurement lag:** {weeks} for signal
- **Total time to impact:** {months}

### Assumptions & Drivers

| Assumption | Conservative | Base Case | Optimistic | Source |
|------------|--------------|-----------|-----------|--------|
| {Segment size} | {X} | {Y} | {Z} | {data source} |
| {Adoption rate} | {X%} | {Y%} | {Z%} | {benchmark or estimate} |
| {ARPU lift} | {$X} | {$Y} | {$Z} | {data or projection} |

### Sensitivity Analysis

What if assumptions change?

Impact under different scenarios:

Best case (all assumptions favor opportunity): ${best}M Base case (reasonable estimates): ${base}M Worst case (conservative assumptions): ${worst}M

Break-even adoption rate: {X%} Break-even ARPU lift: ${X}

### Comparison to Other Opportunities

If multiple opportunities evaluated:

| Opportunity | Annual Impact | Payback | Effort | ROI Rank |
|-------------|---------------|---------|--------|----------|
| {this opportunity} | ${impact} | {months} | {effort} | #{rank} |
| {other opp} | ${impact} | {months} | {effort} | #{rank} |

### Success Metrics & KPIs to Track

| Metric | Baseline | Target | Measurement |
|--------|----------|--------|-------------|
| {metric 1} | {value} | {target} | {how measured} |
| {metric 2} | {value} | {target} | {how measured} |
| {revenue impact} | ${value} | ${target} | Monthly P&L |

### Risks & Assumptions to Validate

| Risk | Impact | Mitigation |
|------|--------|-----------|
| {Adoption lower than expected} | ${lost} | Start with cohort 1, measure before full rollout |
| {Implementation delays} | {months late} | Break into phases, start with MVP |
| {Competitive response} | {market share loss} | Focus on defensible improvements first |

### Recommendation

**Proceed if:**
- This opportunity ranks in the top {N} by ROI
- Key assumptions (adoption, ARPU lift) are validated
- Build effort is estimated at {X} story points or less

**Measure:**
- Weekly tracking of adoption rate and impact metrics
- Monthly cohort analysis to validate assumptions
- Quarterly review of payback and overall business impact

**Next steps:**
1. Validate key assumptions (e.g., survey, prototype test)
2. Design experiment if impact claims are material
3. Build implementation roadmap
4. Brief stakeholders on expected impact

Step 4: Validate Assumptions

Review the sizing with the user:

• Are the addressable user segments realistic?
• Is the adoption rate defensible?
• What assumptions are riskiest and need validation?
• Should we run an experiment before full rollout?

Step 5: Output Sizing Analysis

Save to:

working/opportunity-sizing/{name}_{DATE}.md

Provide:

• Quick summary (total addressable opportunity, annual impact, payback)
• Link to full analysis
• Next actions: "Validate assumptions with a small pilot" or "Ready to prioritize relative to other opportunities"

Templates

Product Feature Template

Addressable = (Total active users) × (will use feature %) × (will upgrade tier %) Impact = Addressable × (incremental ARPU) × (adoption ramp)

Retention/Churn Reduction Template

Addressable = (Churn rate) × (Total users) Impact = Addressable × (save rate %) × (LTV per user)

Efficiency/Cost Reduction Template

Addressable = (Current opex) × (% improvable) Impact = Addressable × (efficiency gain %) Payback = (Implementation cost) / (Annual impact)

Geographic/Segment Expansion Template

Addressable = (New market size) × (penetration %) × (ARPU) Impact = Addressable - (overlap with existing segment)

Edge Cases

• Highly uncertain: Flag confidence as LOW, run scenario analysis
• Long payback period: Consider strategic value beyond financial ROI
• Cannibalization risk: Model as (gross impact) - (lost revenue from existing tiers)
• Cross-team dependencies: Note in risks — can we build independently?

Anti-Patterns

1. Never use a single point estimate — always show range with sensitivity
2. Never ignore adoption ramp — few features reach full penetration overnight
3. Never assume immediate impact — account for measurement lag and ramp time
4. Never forget cannibalization — new tier may shift existing revenue, not add
5. Never confuse addressable with actualized — segment size ≠ captured value