sinapse-ai 1.0.0

package/squad-product/tasks/calculate-product-health-metrics.md

@@ -0,0 +1,162 @@
+---
+task: calculate-product-health-metrics
+responsavel: "@ps-product-analyst"
+responsavel_type: Agent
+atomic_layer: Task
+elicit: false
+
+Entrada:
+  - campo: context
+    tipo: object
+    origem: "workflow ou manual"
+    obrigatorio: true
+
+Saida:
+  - campo: resultado
+    tipo: document
+    destino: "stakeholders"
+
+Checklist:
+  - "[ ] Validar inputs e pre-condicoes"
+  - "[ ] Executar steps conforme documentado"
+  - "[ ] Verificar criterios de qualidade"
+  - "[ ] Gerar output no formato especificado"
+---
+
+# Calculate Product Health Metrics
+
+## Metadata
+- **Agent:** ps-product-analyst (Delta)
+- **Complexity:** Medium
+- **Estimated Time:** 2-3 hours (setup) + weekly updates
+- **Produces:** Product health scorecard, metric calculations, trend analysis, health alerts
+
+## Purpose
+Calcular metricas compostas de saude do produto de forma consistente e reproducivel. Formulas padronizadas para garantir que todos estejam olhando os mesmos numeros.
+
+## Steps
+
+### Step 1: Core Metric Formulas
+
+**Acquisition Metrics:**
+```
+Signup Rate = New Signups / Unique Visitors × 100
+CAC (Customer Acquisition Cost) = Total Marketing Spend / New Customers
+Viral Coefficient (K) = Invitations per User × Conversion Rate of Invitations
+Payback Period = CAC / Monthly Revenue per Customer
+```
+
+**Activation Metrics:**
+```
+Activation Rate = Users who complete Aha Moment / Total Signups × 100
+Time to Value (TTV) = Median(Aha Moment Timestamp - Signup Timestamp)
+Onboarding Completion Rate = Users who complete onboarding / Users who started × 100
+Setup Success Rate = Users with complete setup / Total new users × 100
+```
+
+**Engagement Metrics:**
+```
+DAU/MAU Ratio = Daily Active Users / Monthly Active Users
+  (>25% = excellent engagement, 15-25% = good, <15% = concerning)
+
+Stickiness = DAU / WAU (or DAU/MAU)
+Session Frequency = Total Sessions / Unique Users (per period)
+Session Duration = Avg(session_end - session_start)
+Feature Adoption Rate = Users of Feature X / Total Active Users × 100
+Depth of Engagement = Core Actions per Session (avg)
+```
+
+**Retention Metrics:**
+```
+D7 Retention = Users active on Day 7 / Users who signed up on Day 0 × 100
+D30 Retention = Users active on Day 30 / Users who signed up on Day 0 × 100
+Churn Rate (Monthly) = Customers Lost / Customers at Start of Month × 100
+Net Revenue Retention = (MRR Start + Expansion - Contraction - Churn) / MRR Start × 100
+Logo Retention = (Customers Start - Churned) / Customers Start × 100
+```
+
+**Revenue Metrics:**
+```
+MRR = Sum of all active monthly recurring subscriptions
+ARR = MRR × 12
+ARPU (Average Revenue Per User) = Total Revenue / Total Active Users
+ARPPU (Average Revenue Per Paying User) = Total Revenue / Paying Users
+LTV = ARPU × Average Customer Lifetime (months)
+LTV/CAC Ratio = LTV / CAC (target: >3x)
+Quick Ratio = (New MRR + Expansion MRR) / (Contraction MRR + Churn MRR)
+  (>4 = excellent growth efficiency, 2-4 = good, <2 = leaky bucket)
+```
+
+### Step 2: Product Health Score (Composite)
+**4-Domain Model:**
+```
+Product Health Score = (Product Metrics × 0.30) +
+                       (Delivery Health × 0.25) +
+                       (Client Health × 0.25) +
+                       (Strategic Progress × 0.20)
+```
+
+**Product Metrics Sub-Score (30%):**
+| Metric | Weight | Current | Target | Score (0-100) |
+|--------|--------|---------|--------|--------------|
+| Activation Rate | 25% | [%] | [%] | [0-100] |
+| D30 Retention | 25% | [%] | [%] | [0-100] |
+| Feature Adoption (core) | 20% | [%] | [%] | [0-100] |
+| NPS | 15% | [score] | [score] | [0-100] |
+| Error Rate | 15% | [%] | [<%] | [0-100] |
+
+**Delivery Health Sub-Score (25%):**
+| Metric | Weight | Current | Target | Score (0-100) |
+|--------|--------|---------|--------|--------------|
+| Sprint Velocity Consistency | 30% | [%] | [%] | [0-100] |
+| Deployment Frequency | 25% | [/week] | [/week] | [0-100] |
+| Bug Escape Rate | 25% | [%] | [<%] | [0-100] |
+| Tech Debt Ratio | 20% | [%] | [<%] | [0-100] |
+
+**Client Health Sub-Score (25%):**
+| Metric | Weight | Current | Target | Score (0-100) |
+|--------|--------|---------|--------|--------------|
+| Client Satisfaction (CSAT) | 35% | [score] | [score] | [0-100] |
+| On-time Delivery | 30% | [%] | [%] | [0-100] |
+| Scope Creep | 20% | [%] | [<%] | [0-100] |
+| Communication Quality | 15% | [score] | [score] | [0-100] |
+
+**Strategic Progress Sub-Score (20%):**
+| Metric | Weight | Current | Target | Score (0-100) |
+|--------|--------|---------|--------|--------------|
+| OKR Progress | 40% | [avg score] | [0.6-0.7] | [0-100] |
+| Roadmap Completion | 30% | [%] | [%] | [0-100] |
+| Discovery Velocity | 30% | [experiments/quarter] | [target] | [0-100] |
+
+### Step 3: Health Bands
+| Score Range | Status | Action |
+|------------|--------|--------|
+| 80-100 | GREEN — Healthy | Monitor, optimize |
+| 60-79 | YELLOW — Needs Attention | Investigate, plan improvements |
+| 40-59 | ORANGE — At Risk | Immediate action required |
+| 0-39 | RED — Critical | Escalate, emergency intervention |
+
+### Step 4: Trend Analysis
+**Week-over-Week Movement:**
+| Metric | 4 wks ago | 3 wks ago | 2 wks ago | Last wk | This wk | Trend |
+|--------|----------|----------|----------|---------|---------|-------|
+| Health Score | [N] | [N] | [N] | [N] | [N] | [arrow] |
+| Product | [N] | [N] | [N] | [N] | [N] | [arrow] |
+| Delivery | [N] | [N] | [N] | [N] | [N] | [arrow] |
+| Client | [N] | [N] | [N] | [N] | [N] | [arrow] |
+| Strategic | [N] | [N] | [N] | [N] | [N] | [arrow] |
+
+### Step 5: Alert Rules
+| Condition | Alert Level | Action |
+|-----------|------------|--------|
+| Any domain drops to ORANGE | Warning | Notify product lead |
+| Overall score drops to ORANGE | Urgent | Notify stakeholders + action plan |
+| Any domain drops to RED | Critical | Emergency review meeting |
+| 3 consecutive weeks declining | Trend Alert | Root cause investigation |
+| Metric data missing >24h | Data Alert | Check tracking/pipeline |
+
+## Output Artifacts
+- `product-health-scorecard.md` | `metric-formulas-reference.md` | `health-trend-analysis.md` | `health-alert-log.md`
+
+## Quality Criteria
+- All formulas documented with exact definitions | No ambiguous metric definitions | Health score reproducible (same data = same score) | Trends tracked over min 4 weeks | Alert thresholds defined | Sub-scores traceable to raw metrics

package/squad-product/tasks/calculate-sprint-velocity.md

@@ -0,0 +1,164 @@
+---
+task: calculate-sprint-velocity
+responsavel: "@ps-delivery-manager"
+responsavel_type: Agent
+atomic_layer: Task
+elicit: false
+
+Entrada:
+  - campo: context
+    tipo: object
+    origem: "workflow ou manual"
+    obrigatorio: true
+
+Saida:
+  - campo: resultado
+    tipo: document
+    destino: "stakeholders"
+
+Checklist:
+  - "[ ] Validar inputs e pre-condicoes"
+  - "[ ] Executar steps conforme documentado"
+  - "[ ] Verificar criterios de qualidade"
+  - "[ ] Gerar output no formato especificado"
+---
+
+# Calculate Sprint Velocity
+
+## Metadata
+- **Agent:** ps-delivery-manager (Tempo)
+- **Complexity:** Low-Medium
+- **Estimated Time:** 1-2 hours
+- **Produces:** Velocity report, trend analysis, capacity forecast, predictability metrics
+
+## Purpose
+Calcular e analisar velocidade de sprint para melhorar previsibilidade de delivery. Velocidade nao e meta de performance — e ferramenta de planejamento.
+
+## Steps
+
+### Step 1: Calculate Raw Velocity
+**Sprint Velocity:**
+```
+Velocity = Sum of story points COMPLETED in the sprint
+(Only fully completed items count — partial completion = 0 points)
+
+Sprint [N]: [X] points completed
+Sprint [N-1]: [X] points completed
+Sprint [N-2]: [X] points completed
+
+Rolling Average (last 3 sprints): [avg] points
+Rolling Average (last 6 sprints): [avg] points
+```
+
+**Completion Rate:**
+```
+Completion Rate = Points Completed / Points Committed × 100
+
+Target: 80-90% (consistently >100% means undercommitting)
+```
+
+### Step 2: Velocity Trend Analysis
+**Sprint-over-Sprint:**
+| Sprint | Committed | Completed | Velocity | Completion % | Notes |
+|--------|-----------|-----------|----------|-------------|-------|
+| [N-5] | [pts] | [pts] | [pts] | [%] | [context] |
+| [N-4] | [pts] | [pts] | [pts] | [%] | [context] |
+| [N-3] | [pts] | [pts] | [pts] | [%] | [context] |
+| [N-2] | [pts] | [pts] | [pts] | [%] | [context] |
+| [N-1] | [pts] | [pts] | [pts] | [%] | [context] |
+| [N] | [pts] | [pts] | [pts] | [%] | [context] |
+
+**Trend Assessment:**
+| Pattern | What It Means | Action |
+|---------|-------------|--------|
+| Stable (±10% variation) | Healthy, predictable | Use for planning confidently |
+| Increasing | Team improving or estimates shrinking | Verify: real improvement or estimate inflation? |
+| Decreasing | Impediments, complexity, or team changes | Investigate root cause |
+| Volatile (>20% variation) | Unpredictable, planning is unreliable | Focus on estimation consistency |
+
+### Step 3: Velocity Decomposition
+**Where Did Points Go?**
+| Category | Points | % of Total | Target % |
+|----------|--------|-----------|----------|
+| New features | [pts] | [%] | 60-70% |
+| Tech debt | [pts] | [%] | 15-20% |
+| Bug fixes | [pts] | [%] | 5-10% |
+| Discovery | [pts] | [%] | 10-15% |
+| Unplanned | [pts] | [%] | <10% |
+
+**Unplanned Work Analysis:**
+```
+Unplanned Ratio = Unplanned Points / Total Points × 100
+
+If > 20%: Investigate sources of unplanned work
+Common sources: Production incidents, client urgencies, scope creep, dependencies
+```
+
+### Step 4: Predictability Metrics
+**Velocity Variability:**
+```
+Standard Deviation = σ of last 6 sprint velocities
+Coefficient of Variation = σ / Mean × 100
+
+Predictability Rating:
+  CV < 15%: Highly predictable (GREEN)
+  CV 15-25%: Moderately predictable (YELLOW)
+  CV > 25%: Unpredictable (RED — focus on stabilization)
+```
+
+**Forecast Ranges:**
+```
+Optimistic (P10): [mean + 1.28σ] points per sprint
+Expected (P50): [mean] points per sprint
+Conservative (P80): [mean - 0.84σ] points per sprint
+Pessimistic (P95): [mean - 1.65σ] points per sprint
+
+For N sprints ahead:
+  Expected delivery: N × P50 = [X] points
+  Range: [N × P95] to [N × P10] points
+```
+
+### Step 5: Capacity Adjustments
+**Factors Affecting Next Sprint Velocity:**
+| Factor | Impact | Adjusted Velocity |
+|--------|--------|------------------|
+| Team member on vacation | -[X] pts (proportional) | [adjusted] |
+| New team member (onboarding) | -50% of their capacity | [adjusted] |
+| Holidays in sprint | -[X] pts per day lost | [adjusted] |
+| High unplanned work expected | -15-20% buffer | [adjusted] |
+| Complex/unknown work | -10-15% uncertainty buffer | [adjusted] |
+
+**Next Sprint Capacity Forecast:**
+```
+Base velocity (3-sprint avg): [X] points
+Adjustments: [detail]
+Forecast velocity: [X] points
+Recommended commitment: [90% of forecast] = [X] points
+```
+
+### Step 6: Report & Recommendations
+**Velocity Report:**
+```
+SPRINT VELOCITY REPORT — Sprint [N]
+────────────────────────────────────
+Velocity This Sprint: [X] points
+Rolling Average (3): [X] points
+Completion Rate: [%]
+Predictability: [CV%] — [GREEN/YELLOW/RED]
+
+TREND: [Stable / Improving / Declining / Volatile]
+DECOMPOSITION: Features [%] | Debt [%] | Bugs [%] | Discovery [%] | Unplanned [%]
+
+NEXT SPRINT FORECAST:
+  Recommended commitment: [X] points
+  Range: [low] — [high] points
+
+RECOMMENDATIONS:
+- [specific recommendation based on analysis]
+```
+
+## Output Artifacts
+- `velocity-report-sprint-[N].md` | `velocity-trend-analysis.md` | `capacity-forecast.md`
+
+## Quality Criteria
+- Only completed items counted | Rolling average over 3+ sprints | Trend identified and contextualized | Decomposition by work type | Predictability (CV) calculated | Next sprint forecast with range | Recommendations actionable

package/squad-product/tasks/compute-plg-health-metrics.md

@@ -0,0 +1,194 @@
+---
+task: compute-plg-health-metrics
+responsavel: "@ps-product-analyst"
+responsavel_type: Agent
+atomic_layer: Task
+elicit: false
+
+Entrada:
+  - campo: context
+    tipo: object
+    origem: "workflow ou manual"
+    obrigatorio: true
+
+Saida:
+  - campo: resultado
+    tipo: document
+    destino: "stakeholders"
+
+Checklist:
+  - "[ ] Validar inputs e pre-condicoes"
+  - "[ ] Executar steps conforme documentado"
+  - "[ ] Verificar criterios de qualidade"
+  - "[ ] Gerar output no formato especificado"
+---
+
+# Compute PLG Health Metrics
+
+## Metadata
+- **Agent:** ps-product-analyst (Delta)
+- **Complexity:** Medium-High
+- **Estimated Time:** 3-4 hours
+- **Produces:** PLG scorecard, growth loop analysis, self-serve metrics, product-qualified lead scoring
+
+## Purpose
+Calcular metricas especificas de Product-Led Growth (PLG) para produtos onde o produto e o principal canal de aquisicao, ativacao e monetizacao. Referencia: Wes Bush (Product-Led Growth), Elena Verna, Kyle Poyar (OpenView).
+
+## Steps
+
+### Step 1: PLG Foundation Metrics
+**Self-Serve Funnel:**
+```
+VISITOR → SIGNUP → ACTIVATED → ENGAGED → CONVERTED → EXPANDED
+
+Visitor-to-Signup Rate: [%]
+Signup-to-Activated Rate: [%]
+Activated-to-Engaged Rate: [%]
+Engaged-to-Paid Rate: [%]
+Paid-to-Expanded Rate: [%]
+Overall Visitor-to-Paid: [%]
+```
+
+**Time-Based Metrics:**
+```
+Time to Signup: [median from first visit]
+Time to Value (TTV): [median from signup to aha moment]
+Time to Paid: [median from signup to first payment]
+Free Trial Duration Used: [% of trial period consumed before conversion]
+```
+
+### Step 2: Product-Qualified Lead (PQL) Scoring
+**PQL Definition:**
+```
+A user becomes a PQL when they:
+  1. Complete [activation milestone] AND
+  2. Reach [engagement threshold] AND
+  3. Match [ideal customer profile criteria]
+
+Scoring Model:
+  Behavioral Score (0-100):
+    + [points] for completing onboarding
+    + [points] for using core feature N times
+    + [points] for inviting team members
+    + [points] for reaching usage threshold
+    + [points] for visiting pricing page
+    - [points] for inactivity periods
+
+  Firmographic Score (0-100):
+    + [points] for company size match
+    + [points] for industry match
+    + [points] for role match
+    + [points] for geography match
+
+  PQL Score = (Behavioral × 0.7) + (Firmographic × 0.3)
+  PQL Threshold: Score >= [threshold]
+```
+
+**PQL Metrics:**
+```
+PQL Rate = PQLs / Total Signups × 100
+PQL-to-Customer Rate = Customers from PQLs / Total PQLs × 100
+Time to PQL = Median(PQL timestamp - Signup timestamp)
+PQL Accuracy = Actual conversions from PQLs / Predicted conversions
+```
+
+### Step 3: Growth Loop Metrics
+**Viral Loop:**
+```
+Viral Coefficient (K-Factor):
+  K = i × c
+  Where: i = invitations per user, c = conversion rate per invitation
+  K > 1 = viral growth (each user brings more than 1 new user)
+  K = 0.5-1.0 = supported growth (virality assists other channels)
+  K < 0.5 = minimal virality
+
+Viral Cycle Time: [avg days from user activation to their invite converting]
+Organic Signup %: [% of signups from non-paid channels]
+Word-of-Mouth Index: [NPS promoters × sharing rate]
+```
+
+**Content Loop:**
+```
+User-Generated Content Rate: [content items per active user per month]
+Content Discovery Rate: [% of new visitors from user-generated content]
+SEO Contribution from UGC: [organic traffic from user content pages]
+```
+
+**Usage Loop:**
+```
+Network Effect Strength: [value increase per additional user/node]
+Data Network Effect: [product improvement rate per data point added]
+Switching Cost Index: [integrations + data + habits + team usage]
+```
+
+### Step 4: Monetization Efficiency
+**PLG-Specific Revenue Metrics:**
+```
+Natural Rate of Revenue (NRR):
+  Revenue from self-serve conversions (no sales touch) / Total new revenue × 100
+  Target: >50% for PLG company
+
+Sales-Assist Rate:
+  Revenue where sales accelerated (but user started self-serve) / Total new revenue × 100
+
+Average Contract Value by Motion:
+  Self-serve ACV: [$]
+  Sales-assist ACV: [$]
+  Sales-led ACV: [$]
+
+Expansion Revenue Rate:
+  Revenue from upgrades + add-ons / Total revenue × 100
+  PLG target: >30%
+
+Product-Driven Expansion Rate:
+  Expansion triggered by usage limits / Total expansion × 100
+```
+
+**Quick Ratio (Growth Efficiency):**
+```
+Quick Ratio = (New MRR + Expansion MRR) / (Contraction MRR + Churn MRR)
+
+Interpretation:
+  > 4.0 = Excellent (hyper-growth efficient)
+  2.0-4.0 = Good (healthy growth)
+  1.0-2.0 = Concerning (leaky bucket)
+  < 1.0 = Shrinking (losing more than gaining)
+```
+
+### Step 5: PLG Health Scorecard
+**Composite PLG Score:**
+| Dimension | Metric | Weight | Current | Target | Score (0-100) |
+|-----------|--------|--------|---------|--------|--------------|
+| Self-Serve | Visitor-to-Paid rate | 15% | [%] | [%] | |
+| Activation | Time to Value | 15% | [days] | [days] | |
+| PQL | PQL-to-Customer rate | 15% | [%] | [%] | |
+| Virality | K-Factor | 10% | [score] | [target] | |
+| Engagement | DAU/MAU | 10% | [ratio] | [ratio] | |
+| Retention | D30 retention | 10% | [%] | [%] | |
+| Monetization | Natural Rate of Revenue | 10% | [%] | [%] | |
+| Expansion | Product-Driven Expansion | 10% | [%] | [%] | |
+| Efficiency | Quick Ratio | 5% | [ratio] | [ratio] | |
+| **TOTAL** | | **100%** | | | **/100** |
+
+**PLG Maturity Bands:**
+| Score | Stage | Description |
+|-------|-------|-------------|
+| 80-100 | PLG-Native | Product is the primary growth engine |
+| 60-79 | PLG-Supported | Product assists but sales still drives |
+| 40-59 | PLG-Emerging | Self-serve exists but underperforms |
+| 0-39 | Sales-Led | Minimal product-led motion |
+
+### Step 6: Recommendations by PLG Stage
+**Based on Current Score:**
+| Stage | Top Priority | Key Investment |
+|-------|-------------|---------------|
+| PLG-Native | Optimize expansion loops | Usage-based pricing, network effects |
+| PLG-Supported | Reduce TTV, improve PQL accuracy | Onboarding, activation, PQL model |
+| PLG-Emerging | Build self-serve funnel end-to-end | Pricing page, trial flow, activation |
+| Sales-Led | Validate PLG viability | Free tier/trial, tracking, activation |
+
+## Output Artifacts
+- `plg-health-scorecard.md` | `pql-scoring-model.md` | `growth-loop-analysis.md` | `plg-recommendations.md`
+
+## Quality Criteria
+- Self-serve funnel fully instrumented | PQL model defined with scoring rules | Growth loops identified and measured | Monetization split by motion (self-serve vs sales) | PLG maturity honestly assessed | Recommendations match current stage

package/squad-product/tasks/conduct-feature-adoption-audit.md

@@ -0,0 +1,145 @@
+---
+task: conduct-feature-adoption-audit
+responsavel: "@ps-product-analyst"
+responsavel_type: Agent
+atomic_layer: Task
+elicit: false
+
+Entrada:
+  - campo: context
+    tipo: object
+    origem: "workflow ou manual"
+    obrigatorio: true
+
+Saida:
+  - campo: resultado
+    tipo: document
+    destino: "stakeholders"
+
+Checklist:
+  - "[ ] Validar inputs e pre-condicoes"
+  - "[ ] Executar steps conforme documentado"
+  - "[ ] Verificar criterios de qualidade"
+  - "[ ] Gerar output no formato especificado"
+---
+
+# Conduct Feature Adoption Audit
+
+## Metadata
+- **Agent:** ps-product-analyst (Delta)
+- **Complexity:** Medium
+- **Estimated Time:** 3-4 hours
+- **Produces:** Feature adoption matrix, usage patterns, optimization recommendations, sunset candidates
+
+## Purpose
+Auditar adocao de features para entender quais features geram valor, quais estao subutilizadas e quais sao candidatas a sunset. Base para decisoes de investimento em produto.
+
+## Steps
+
+### Step 1: Inventory All Features
+**Feature Catalog:**
+| Feature | Category | Launch Date | Target Persona | Maintenance Cost |
+|---------|----------|-----------|---------------|-----------------|
+| [feature 1] | Core/Secondary/Peripheral | [date] | [persona] | Low/Med/High |
+| [feature 2] | | | | |
+| [feature 3] | | | | |
+
+### Step 2: Measure Adoption Metrics
+**Per Feature:**
+```
+FEATURE: [name]
+──────────────────────────────
+Discovery Rate: [% of users who see/find the feature]
+Adoption Rate: [% of users who use it at least once]
+Activation Rate: [% who use it and get value (complete task)]
+Retention Rate: [% who use it repeatedly]
+Depth of Use: [avg actions per session within feature]
+Time to First Use: [median days from signup to first use]
+```
+
+**Feature Adoption Matrix (2x2):**
+```
+                    HIGH BREADTH (many users)
+                    │
+    CORE            │           POWER
+    (High breadth   │     (High breadth +
+     High depth)    │      High depth)
+                    │
+────────────────────┼────────────────────
+                    │
+    UNDERUSED       │           ZOMBIE
+    (Low breadth    │     (Low breadth
+     Low depth)     │      Low depth)
+                    │
+                    LOW BREADTH (few users)
+```
+
+**Quadrant Definitions:**
+| Quadrant | Adoption | Depth | Strategy |
+|----------|----------|-------|----------|
+| **Core** | >50% users | >3x/week | Protect and optimize |
+| **Power** | >30% users | >5x/week | Reduce friction, promote |
+| **Underused** | <20% users | <1x/week | Improve discovery or sunset |
+| **Zombie** | <5% users | <1x/month | Strong sunset candidate |
+
+### Step 3: Analyze Adoption Patterns
+**Adoption by Segment:**
+| Feature | Persona A | Persona B | Persona C | Free | Paid |
+|---------|-----------|-----------|-----------|------|------|
+| [feature] | [%] | [%] | [%] | [%] | [%] |
+
+**Correlation with Key Outcomes:**
+| Feature Usage | Retention Impact | Revenue Impact | NPS Impact |
+|--------------|-----------------|---------------|-----------|
+| [feature 1] | +[X]pp retention | +$[Y] ARPU | +[Z] NPS |
+| [feature 2] | +[X]pp | +$[Y] | +[Z] |
+
+**Feature Discovery Analysis:**
+| Feature | Discovery Method | Discovery Rate | Adoption After Discovery |
+|---------|-----------------|---------------|------------------------|
+| [feature] | [navigation/search/onboarding/referral] | [%] | [%] |
+
+### Step 4: Identify Issues
+**Low Adoption Root Causes:**
+| Feature | Adoption Rate | Likely Root Cause | Evidence |
+|---------|-------------|------------------|---------|
+| [feature] | [%] | Discovery problem (users do not find it) | [low discovery rate] |
+| [feature] | [%] | Value gap (users do not see benefit) | [high discovery, low adoption] |
+| [feature] | [%] | Usability issue (too hard to use) | [high adoption, low activation] |
+| [feature] | [%] | Wrong audience (built for wrong persona) | [segment mismatch] |
+| [feature] | [%] | Timing issue (offered too early/late) | [time-to-use pattern] |
+
+### Step 5: Generate Recommendations
+**Per Feature:**
+| Feature | Quadrant | Action | Rationale | Priority |
+|---------|---------|--------|-----------|----------|
+| [feature] | Core | Optimize performance | High-usage, high-value | P1 |
+| [feature] | Power | Improve discovery | Valuable but hidden | P1 |
+| [feature] | Underused | A/B test discovery | May have discovery problem | P2 |
+| [feature] | Zombie | Evaluate for sunset | Low value, maintenance cost | P2 |
+
+**Sunset Evaluation (for Zombie features):**
+| Criterion | Assessment |
+|-----------|-----------|
+| Current active users | [N absolute users] |
+| Revenue from these users | [$] |
+| Maintenance cost per quarter | [$] |
+| Dependencies (other features that use this) | [list] |
+| Migration path for current users | [what they would use instead] |
+| Recommendation | Sunset / Rethink / Keep (with justification) |
+
+### Step 6: Compile Audit Report
+**Structure:**
+1. Feature landscape overview (total features, distribution across quadrants)
+2. Feature adoption matrix (visual 2x2)
+3. Top findings (high-impact insights)
+4. Feature-by-feature analysis (detailed per feature)
+5. Recommendations (prioritized action plan)
+6. Sunset candidates (with migration paths)
+7. Investment allocation suggestion
+
+## Output Artifacts
+- `feature-adoption-audit.md` | `feature-matrix.md` | `sunset-candidates.md` | `feature-optimization-plan.md`
+
+## Quality Criteria
+- All features inventoried | Adoption metrics calculated per feature | 2x2 matrix populated | Segment analysis completed | Root causes identified for low adoption | Sunset candidates evaluated with criteria | Recommendations prioritized by impact