sinapse-ai 1.0.0

package/squad-product/tasks/map-current-state-journey.md

@@ -0,0 +1,134 @@
+---
+task: map-current-state-journey
+responsavel: "@ps-discovery-lead"
+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"
+---
+
+# Map Current-State Journey
+
+## Metadata
+- **Agent:** ps-discovery-lead (Quorum)
+- **Complexity:** Medium-High
+- **Estimated Time:** 4-6 hours
+- **Produces:** Current-state journey map, pain point inventory, opportunity map
+
+## Purpose
+Mapear a jornada atual do usuario (as-is) para identificar pain points, momentos de friccao e oportunidades de melhoria. Baseado em dados reais de comportamento, nao em suposicoes.
+
+## Steps
+
+### Step 1: Define Journey Scope
+**Boundary Definition:**
+```
+Persona: [which persona this journey maps]
+Journey: [name — e.g., "First purchase", "Onboarding", "Report creation"]
+Start Point: [trigger that initiates the journey]
+End Point: [successful outcome or abandonment]
+Channels: [web, mobile, email, support, etc.]
+```
+
+**Scope Rules:**
+- Map ONE journey at a time (not the entire product)
+- Focus on the most impactful journey first (highest traffic OR highest drop-off)
+- Include pre-product and post-product touchpoints
+
+### Step 2: Collect Journey Data
+**Data Sources:**
+| Source | What It Reveals | Collection Method |
+|--------|----------------|-------------------|
+| Analytics (funnels) | Where users drop off | Funnel analysis in analytics tool |
+| Session recordings | How users actually behave | Hotjar/FullStory review (20+ sessions) |
+| User interviews | Why they do what they do | Discovery interview excerpts |
+| Support tickets | Where they get stuck | Ticket categorization by journey stage |
+| NPS/CSAT by touchpoint | Satisfaction per stage | Survey data segmented by journey point |
+| Heatmaps | Where attention goes | Click/scroll heatmap analysis |
+
+**Minimum Data for Valid Journey Map:**
+- [ ] Funnel analytics for the journey (quantitative flow)
+- [ ] 5+ user interviews covering this journey
+- [ ] 20+ session recordings reviewed
+- [ ] Support ticket themes categorized
+
+### Step 3: Map Journey Stages
+**Per Stage:**
+```
+┌─────────────────────────────────────────────────────────────┐
+│ STAGE: [name]                                               │
+├─────────────────────────────────────────────────────────────┤
+│ User Goal: What are they trying to accomplish?              │
+│ Actions: What do they actually DO? (observed, not assumed)  │
+│ Touchpoints: Where does this happen? (screen, email, etc.) │
+│ Thoughts: What are they thinking? (from interviews)         │
+│ Emotions: How do they feel? (😤😐😊) (from interviews)       │
+│ Pain Points: What is hard, confusing, or frustrating?       │
+│ Workarounds: How do they compensate for pain points?        │
+│ Drop-off Rate: % who leave at this stage (from analytics)   │
+│ Time Spent: Average duration at this stage                  │
+│ Support Tickets: Volume related to this stage               │
+└─────────────────────────────────────────────────────────────┘
+```
+
+### Step 4: Identify Moments of Truth
+**Critical Moments:**
+| Moment Type | Definition | Example |
+|-------------|-----------|---------|
+| **Aha Moment** | When user first realizes value | "Oh, this saves me 2 hours!" |
+| **Friction Point** | Unnecessary difficulty | "Why do I need to fill this again?" |
+| **Anxiety Point** | User feels uncertain/risky | "Will this delete my data?" |
+| **Delight Moment** | Exceeds expectations | "It auto-filled everything!" |
+| **Abandonment Trigger** | User leaves | "This is too complicated" |
+
+### Step 5: Quantify Pain Points
+**Pain Point Severity Score:**
+| Pain Point | Frequency (% users affected) | Severity (1-5) | Impact on Conversion | Score |
+|-----------|------------------------------|----------------|---------------------|-------|
+| [pain 1] | [%] | [1-5] | [drop-off %] | Freq × Severity |
+| [pain 2] | [%] | [1-5] | [drop-off %] | Freq × Severity |
+
+**Priority:** Address highest-score pain points first.
+
+### Step 6: Map Opportunities
+**Per Pain Point, Identify Opportunity:**
+```
+Pain Point: [from journey map]
+Stage: [which journey stage]
+Severity Score: [from Step 5]
+Opportunity: [how might we address this?]
+Type: Remove friction | Add clarity | Automate | Redesign | Eliminate step
+Estimated Impact: [conversion lift or time saved]
+Effort: [T-shirt size: XS/S/M/L/XL]
+```
+
+### Step 7: Compile Journey Map Document
+**Format:**
+1. Journey overview (persona, scope, key metrics)
+2. Stage-by-stage map (visual if possible, text if not)
+3. Emotion curve (high points and low points across stages)
+4. Pain point inventory (prioritized)
+5. Opportunity map (prioritized by impact/effort)
+6. Recommendations (top 3 quick wins + top 3 strategic improvements)
+
+## Output Artifacts
+- `journey-map-[persona]-[journey].md` | `pain-point-inventory.md` | `opportunity-map.md` | `journey-data-sources.md`
+
+## Quality Criteria
+- Based on real data (analytics + interviews + recordings) | All stages mapped with actions + emotions | Pain points quantified with severity scores | Opportunities connected to pain points | Min 20 session recordings reviewed | Drop-off rates included per stage

package/squad-product/tasks/map-opportunity-solution-tree.md

@@ -0,0 +1,154 @@
+---
+task: map-opportunity-solution-tree
+responsavel: "@ps-discovery-lead"
+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"
+---
+
+# Map Opportunity Solution Tree
+
+## Metadata
+- **Agent:** ps-discovery-lead (Quorum)
+- **Complexity:** Medium
+- **Estimated Time:** 2-3 hours (initial) + weekly maintenance
+- **Produces:** Opportunity Solution Tree, prioritized opportunities, experiment backlog
+
+## Purpose
+Criar e manter Opportunity Solution Tree (Teresa Torres) como ferramenta central de discovery. Conecta outcome desejado a oportunidades reais de usuarios e solucoes testadas.
+
+## Steps
+
+### Step 1: Define Desired Outcome (Root)
+**The outcome must be:**
+- [ ] A business metric the team can influence
+- [ ] Specific and measurable
+- [ ] Time-bound
+- [ ] Connected to company OKR or product strategy
+
+**Format:** "[Metric] from [current] to [target] by [date]"
+
+**Examples:**
+- "Increase 30-day retention from 35% to 50% by Q4"
+- "Reduce time-to-first-value from 7 days to 2 days by Q3"
+- "Increase conversion from trial to paid from 5% to 12% by Q2"
+
+### Step 2: Map Opportunities (Level 1 & 2)
+**Sources for Opportunities:**
+- User interviews (pain points, unmet needs)
+- Analytics (drop-off points, low-adoption features)
+- Support data (recurring complaints, feature requests)
+- Journey maps (friction points, moments of truth)
+
+**Opportunity Framing Rules (Teresa Torres):**
+- Write as user needs or pain points, NOT solutions
+- BAD: "Add search feature" (solution, not opportunity)
+- GOOD: "Users cannot find relevant content quickly" (opportunity)
+- Each opportunity should emerge from real user evidence
+
+**Tree Structure:**
+```
+OUTCOME: [metric target]
+├── Opportunity 1: [user need/pain point]
+│   ├── Opportunity 1.1: [more specific aspect]
+│   └── Opportunity 1.2: [more specific aspect]
+├── Opportunity 2: [user need/pain point]
+│   ├── Opportunity 2.1: [more specific aspect]
+│   └── Opportunity 2.2: [more specific aspect]
+└── Opportunity 3: [user need/pain point]
+    ├── Opportunity 3.1: [more specific aspect]
+    └── Opportunity 3.2: [more specific aspect]
+```
+
+### Step 3: Prioritize Opportunities
+**Assessment per Opportunity:**
+| Opportunity | Evidence Strength | Market Size | Strategic Fit | Urgency | Score |
+|-------------|------------------|-------------|--------------|---------|-------|
+| [opp 1] | [1-5] | [1-5] | [1-5] | [1-5] | [avg] |
+
+**Evidence Strength Scale:**
+- 5: Multiple data sources, strong triangulation
+- 4: Clear pattern from interviews + some analytics
+- 3: Emerging pattern from interviews only
+- 2: Single data point or assumption
+- 1: Pure hypothesis, no evidence
+
+**Selection Rule:** Focus on TOP 3 opportunities at a time. Do not spread team thin across too many opportunities.
+
+### Step 4: Generate Solutions (per Priority Opportunity)
+**Per Opportunity, brainstorm 3+ solutions:**
+```
+Opportunity: [user need/pain point]
+├── Solution A: [approach 1]
+│   Effort: [T-shirt] | Impact estimate: [High/Med/Low] | Risk: [High/Med/Low]
+├── Solution B: [approach 2]
+│   Effort: [T-shirt] | Impact estimate: [High/Med/Low] | Risk: [High/Med/Low]
+└── Solution C: [approach 3]
+    Effort: [T-shirt] | Impact estimate: [High/Med/Low] | Risk: [High/Med/Low]
+```
+
+**Rules:**
+- Always generate multiple solutions (avoid first-idea bias)
+- Include at least one "simple/cheap" option
+- Solutions should be specific enough to test
+
+### Step 5: Design Experiments (per Solution)
+**Experiment Ladder (cheapest first):**
+1. Assumption test (survey, desk research)
+2. Smoke test (fake door, landing page)
+3. Wizard of Oz (manual backend, real frontend)
+4. Concierge (do it manually for users)
+5. Prototype test (clickable prototype)
+6. Alpha/beta launch (limited rollout)
+7. Full launch (measure in production)
+
+**Per Experiment:**
+```
+Solution: [which solution this tests]
+Assumption: [what we need to learn]
+Method: [experiment type from ladder]
+Success criteria: [specific threshold]
+Duration: [days/weeks]
+Cost: [effort needed]
+```
+
+### Step 6: Maintain the Tree (Weekly)
+**Weekly OST Review (15 min):**
+- [ ] Update with new interview insights
+- [ ] Add new opportunities discovered
+- [ ] Move experiments forward (results → decisions)
+- [ ] Prune dead-end opportunities (invalidated)
+- [ ] Ensure top 3 opportunities have active experiments
+- [ ] Check: Are we making progress toward the outcome?
+
+**Health Check:**
+| Indicator | Healthy | Unhealthy |
+|-----------|---------|-----------|
+| Active opportunities | 2-4 in focus | 0 or >6 |
+| Solutions per opportunity | 2-4 options | 1 (no exploration) |
+| Experiments running | 1-3 active | 0 (stalled) |
+| Tree freshness | Updated weekly | Stale >2 weeks |
+| Evidence per opportunity | 2+ sources | Assumptions only |
+
+## Output Artifacts
+- `opportunity-solution-tree.md` | `opportunity-priority-matrix.md` | `experiment-backlog.md` | `ost-weekly-log.md`
+
+## Quality Criteria
+- Outcome measurable and time-bound | Opportunities based on evidence, not assumptions | Min 3 solutions per priority opportunity | Experiments start from cheapest | Weekly maintenance active | Top 3 opportunities always have active work

package/squad-product/tasks/model-churn-prediction.md

@@ -0,0 +1,174 @@
+---
+task: model-churn-prediction
+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"
+---
+
+# Model Churn Prediction
+
+## Metadata
+- **Agent:** ps-product-analyst (Delta)
+- **Complexity:** High
+- **Estimated Time:** 4-6 hours
+- **Produces:** Churn prediction model, risk scores, early warning signals, intervention playbook
+
+## Purpose
+Construir modelo preditivo de churn para identificar usuarios em risco ANTES de cancelarem. Foco em sinais comportamentais acionaveis, nao apenas correlacoes estatisticas.
+
+## Steps
+
+### Step 1: Define Churn
+**Churn Definition (be specific):**
+```
+A user is considered churned when:
+  [specific behavioral definition]
+
+Examples:
+  - SaaS: No login for 30 consecutive days
+  - Subscription: Subscription cancelled or not renewed
+  - Freemium: No core action for 14 days
+  - Agency client: Contract not renewed at end of term
+```
+
+**Churn Types to Track:**
+| Type | Definition | Prediction Window |
+|------|-----------|------------------|
+| Voluntary churn | User actively cancels | 30-60 days ahead |
+| Involuntary churn | Payment failure, no renewal | 7-14 days ahead |
+| Silent churn | Gradual disengagement, stops using | 14-30 days ahead |
+| Downgrade | Moves to lower plan | 30 days ahead |
+
+### Step 2: Identify Leading Indicators
+**Behavioral Signals (from analytics):**
+| Signal | Typical Pattern | Weight |
+|--------|----------------|--------|
+| Login frequency decline | 50%+ drop vs previous period | High |
+| Core feature usage decline | Stopped using primary feature | High |
+| Session duration decline | Shorter sessions over time | Medium |
+| Support ticket spike | 3+ tickets in a week | Medium |
+| Feature exploration stops | Only uses 1-2 features | Medium |
+| Export/download spike | Extracting data before leaving | High |
+| Billing page visits | Checking plan/cancellation | Very High |
+| Invite/sharing stops | No collaborative activity | Medium |
+| Error encounter rate | Frequent errors without resolution | Medium |
+
+**Engagement Score (composite):**
+```
+Engagement Score = w1 × (login frequency normalized) +
+                   w2 × (core feature usage normalized) +
+                   w3 × (feature breadth normalized) +
+                   w4 × (session depth normalized) +
+                   w5 × (collaboration activity normalized)
+
+Score range: 0-100
+Risk bands: >70 = Safe | 40-70 = Monitor | <40 = At Risk
+```
+
+### Step 3: Build Risk Scoring Model
+**Simple Heuristic Model (start here):**
+| Risk Factor | Points | Condition |
+|------------|--------|-----------|
+| Login decline >50% | +30 | vs 30-day average |
+| No core action 7+ days | +25 | Feature X not used |
+| Support tickets 3+ (week) | +15 | Recent week |
+| NPS detractor | +15 | Last survey score 0-6 |
+| Billing page visit | +20 | In last 14 days |
+| Session duration decline >40% | +10 | vs 30-day average |
+| Error encounters +3 (week) | +10 | Unresolved errors |
+| No feature exploration | +5 | Same features only |
+| **TOTAL POSSIBLE** | **130** | |
+
+**Risk Tiers:**
+| Score | Risk Level | % Likely to Churn | Action |
+|-------|-----------|-------------------|--------|
+| 0-25 | Low | <5% | Monitor |
+| 26-50 | Medium | 10-25% | Proactive outreach |
+| 51-80 | High | 25-50% | Urgent intervention |
+| 81+ | Critical | >50% | Emergency save attempt |
+
+### Step 4: Validate Model (Backtest)
+**Validation Method:**
+1. Take 3-6 months of historical data
+2. Calculate risk scores for each user at each point in time
+3. Compare predicted risk with actual churn outcome
+4. Calculate model accuracy:
+
+```
+Precision = True Positives / (True Positives + False Positives)
+  "Of those we flagged as at-risk, how many actually churned?"
+
+Recall = True Positives / (True Positives + False Negatives)
+  "Of those who churned, how many did we flag?"
+
+F1 Score = 2 × (Precision × Recall) / (Precision + Recall)
+  Target: >0.70
+```
+
+**Confusion Matrix:**
+```
+                  Predicted: Churn    Predicted: Stay
+Actual: Churned    [True Positive]    [False Negative]
+Actual: Stayed     [False Positive]    [True Negative]
+```
+
+### Step 5: Design Intervention Playbook
+**Per Risk Level:**
+```
+RISK LEVEL: [Medium / High / Critical]
+─────────────────────────────────────
+Trigger: Score reaches [threshold]
+Timeline: Intervene within [N days]
+
+INTERVENTION SEQUENCE:
+1. [Day 0] Automated: [email/in-app message]
+   Content: [personalized based on risk factor]
+2. [Day 3] Personal: [CS outreach]
+   Script: [what to say/ask]
+3. [Day 7] Escalation: [manager/product lead involvement]
+   Offer: [retention offer if applicable]
+4. [Day 14] Final: [exit interview / save attempt]
+
+SUCCESS CRITERIA: User engagement score returns to >70 within 30 days
+TRACKING: Log intervention + outcome for model improvement
+```
+
+### Step 6: Monitoring & Iteration
+**Weekly Churn Dashboard:**
+| Metric | This Week | Last Week | Trend |
+|--------|----------|----------|-------|
+| Users at Critical risk | [N] | [N] | [direction] |
+| Users at High risk | [N] | [N] | [direction] |
+| Interventions triggered | [N] | [N] | |
+| Saves (returned to Low risk) | [N] | [N] | |
+| Actual churns | [N] | [N] | |
+| Model accuracy (rolling 30d) | [%] | [%] | |
+
+**Model Improvement Cycle:**
+- Monthly: Review false positives and false negatives
+- Quarterly: Retrain weights based on new data
+- Continuous: Add new signals as product evolves
+
+## Output Artifacts
+- `churn-prediction-model.md` | `risk-scoring-rules.md` | `intervention-playbook.md` | `model-validation-results.md` | `churn-dashboard-spec.md`
+
+## Quality Criteria
+- Churn clearly defined | Leading indicators identified from data (not assumed) | Risk score model documented with weights | Backtested against historical data | F1 score > 0.70 | Intervention playbook per risk level | Monthly review process defined

package/squad-product/tasks/negotiate-scope-change-request.md

@@ -0,0 +1,185 @@
+---
+task: negotiate-scope-change-request
+responsavel: "@ps-client-product-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"
+---
+
+# Negotiate Scope Change Request
+
+## Metadata
+- **Agent:** ps-client-product-manager (Proxy)
+- **Complexity:** Medium
+- **Estimated Time:** 1-2 hours
+- **Produces:** Scope change assessment, trade-off options, updated agreement, communication
+
+## Purpose
+Gerenciar mudancas de escopo de forma profissional e transparente, protegendo a equipe de scope creep enquanto mantendo a relacao com o cliente positiva. Escopo muda — a questao e COMO gerenciar a mudanca.
+
+## Steps
+
+### Step 1: Document the Change Request
+**Change Request Form:**
+```
+CHANGE REQUEST: CR-[ID]
+Date: [date]
+Client: [client name]
+Requester: [person + role]
+Priority: [URGENT/HIGH/MEDIUM/LOW]
+
+CURRENT SCOPE: [what was agreed]
+REQUESTED CHANGE: [what client wants to add/modify/remove]
+REASON FOR CHANGE: [why — business context]
+```
+
+### Step 2: Impact Assessment
+**Effort Impact:**
+| Dimension | Estimate |
+|-----------|---------|
+| Additional effort | [T-shirt / story points / days] |
+| Additional cost | [$ or hours × rate] |
+| Timeline impact | [N days/weeks added] |
+| Team impact | [who is affected, availability] |
+
+**Scope Impact:**
+| Aspect | Before Change | After Change |
+|--------|-------------|-------------|
+| Total scope (points/hours) | [N] | [N + delta] |
+| Sprint/cycle end date | [date] | [new date if delayed] |
+| Quality risk | [level] | [new level if rushed] |
+| Other features affected | [none/list] | [what gets impacted] |
+
+**Risk Assessment:**
+| Risk | Probability | Impact | Mitigation |
+|------|------------|--------|-----------|
+| Delivery delay | [%] | [description] | [how to manage] |
+| Quality compromise | [%] | [description] | [how to manage] |
+| Team burnout | [%] | [description] | [how to manage] |
+| Budget overrun | [%] | [description] | [how to manage] |
+
+### Step 3: Generate Trade-off Options
+**Always present options, never just "yes" or "no":**
+
+**Option A: Accept and Extend**
+```
+Accept change + extend timeline by [N days/weeks]
+Cost: [additional $]
+Timeline: New deadline [date]
+Risk: Low (clean approach)
+```
+
+**Option B: Accept and Trade**
+```
+Accept change BUT remove/defer [item X] from scope
+Cost: Neutral (within original budget)
+Timeline: Maintained
+Risk: Medium (what we lose by deferring X)
+Trade: "[Added item] replaces [removed item]"
+```
+
+**Option C: Accept and Compress**
+```
+Accept change within original timeline
+Cost: [additional $ for more resources or overtime]
+Timeline: Maintained
+Risk: HIGH (quality risk, team fatigue)
+Note: Not recommended if quality is non-negotiable
+```
+
+**Option D: Defer to Next Phase**
+```
+Log the request for next sprint/quarter
+Cost: None now
+Timeline: Maintained for current scope
+Risk: Low (but client may be disappointed)
+Benefit: Proper planning, no disruption
+```
+
+### Step 4: Recommendation
+**Decision Template:**
+```
+RECOMMENDATION: [Option A/B/C/D]
+
+RATIONALE:
+[Why this option is best for client AND for delivery quality]
+
+TRADE-OFFS:
+[What we gain and what we sacrifice with this choice]
+
+IF CLIENT DISAGREES:
+[Fallback option + what that would mean]
+```
+
+### Step 5: Client Communication
+**Scope Change Discussion Script:**
+```
+1. ACKNOWLEDGE: "I understand this is important because [their reason]."
+2. ASSESS: "Let me share what this would involve..."
+3. OPTIONS: "I see [N] ways we could handle this..."
+4. RECOMMEND: "I recommend [option] because..."
+5. DECIDE: "What would you prefer?"
+6. DOCUMENT: "Let me document this decision for our records."
+```
+
+**Communication Principles:**
+- Never say "no" — always offer alternatives
+- Frame as trade-offs, not limitations
+- Use data (effort, timeline) not emotions
+- Let the client choose (with your recommendation)
+- Document everything (protects both parties)
+
+### Step 6: Formalize Change
+**Change Order (if contract requires):**
+```
+CHANGE ORDER: CO-[ID]
+Related to CR: [change request ID]
+Date Approved: [date]
+Approved by: [client name + agency lead]
+
+CHANGE DESCRIPTION:
+[Specific scope change]
+
+IMPACT:
+  Additional effort: [hours/points]
+  Additional cost: [$ — if billable]
+  Timeline change: [new milestone dates]
+
+ITEMS TRADED (if Option B):
+  Added: [new items]
+  Removed/Deferred: [traded items]
+
+SIGNATURES:
+  Client: _____________ Date: _____
+  Agency: _____________ Date: _____
+```
+
+**Update Tracking:**
+- [ ] Backlog updated with new/modified items
+- [ ] Timeline/roadmap updated
+- [ ] Team informed of change
+- [ ] Change logged in decision log
+- [ ] Contract/SOW amended if needed
+
+## Output Artifacts
+- `change-request-[id].md` | `scope-impact-assessment.md` | `change-order-[id].md` | `scope-change-log.md`
+
+## Quality Criteria
+- Change documented before acting | Impact quantified (effort + timeline + cost) | Multiple options presented | Recommendation clear with rationale | Client chose (not forced) | Decision documented formally | Team informed