sinapse-ai 1.0.0

package/squad-growth/tasks/design-onboarding-flow.md

@@ -0,0 +1,162 @@
+---
+task: design-onboarding-flow
+responsavel: "@ga-growth-hacker"
+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"
+---
+
+# Task: Design Onboarding Flow
+
+## Metadata
+- **Squad:** squad-growth
+- **Agent:** Lever (ga-growth-hacker)
+- **Complexity:** Standard
+
+## Objetivo
+Desenhar onboarding flow — criar experiencia de primeiro uso que guia o usuario do signup ao primeiro valor, maximizando ativacao e criando habito de uso.
+
+## Entrada
+- User personas and goals
+- Core product value proposition
+- Current onboarding flow (if exists)
+- Activation data and benchmarks
+
+## Passos
+
+### 1. Onboarding Patterns
+| Pattern | Descricao | Melhor Para | Exemplo |
+|---------|----------|------------|---------|
+| Product Tour | Walkthrough guiado do produto | Complex SaaS | Notion, Figma |
+| Progressive Disclosure | Revela features gradualmente | Feature-rich products | Slack |
+| Task-Based | "Complete essas 3 tarefas" | Goal-oriented products | Asana |
+| Wizard | Step-by-step setup obrigatorio | Config-heavy products | HubSpot |
+| Learn by Doing | Sandbox com dados de exemplo | Creative tools | Canva |
+| Video Walkthrough | Video curto mostrando valor | Visual products | Loom |
+| Empty State CTA | Guide via empty states | Simple products | Todoist |
+
+### 2. Onboarding Flow Structure
+```
+┌────────────────────────────────────────┐
+│ PHASE 1: SIGNUP (minimize friction)    │
+│ • Social login (Google/Apple)          │
+│ • Email + password only               │
+│ • Skip everything non-essential        │
+│              │                         │
+│              ▼                         │
+│ PHASE 2: PERSONALIZATION (1-3 questions)│
+│ • "What's your role?"                  │
+│ • "What's your main goal?"             │
+│ • "How big is your team?"              │
+│ → Customize experience based on answers│
+│              │                         │
+│              ▼                         │
+│ PHASE 3: QUICK WIN (< 2 minutes)      │
+│ • Template selection or sample data    │
+│ • Guide to first core action           │
+│ • Celebrate completion!                │
+│              │                         │
+│              ▼                         │
+│ PHASE 4: DEEPER VALUE (next session)   │
+│ • Advanced features introduction       │
+│ • Team invite prompt                   │
+│ • Integration setup                    │
+│              │                         │
+│              ▼                         │
+│ PHASE 5: HABIT (first 2 weeks)         │
+│ • Streak/progress tracking             │
+│ • Contextual tips per feature          │
+│ • Milestone celebrations               │
+└────────────────────────────────────────┘
+```
+
+### 3. Onboarding Checklist Design
+| Item | Descricao | Reward | Priority |
+|------|----------|--------|---------|
+| Complete profile | Foto + bio | Personalized experience | P1 |
+| First core action | Criar/fazer primeiro item | Celebracao + badge | P0 |
+| Explore feature X | Usar feature secundaria | Tooltip unlock | P2 |
+| Invite teammate | Convidar 1 pessoa | Extended trial / bonus | P1 |
+| Connect integration | Integrar ferramenta externa | Workflow automation | P2 |
+
+**Design rules:**
+- Maximum 5 items in checklist
+- Most important item first
+- Show progress (3/5 complete)
+- Allow skip but encourage completion
+
+### 4. Personalization Branches
+| Resposta | Path | Conteudo Adaptado |
+|---------|------|------------------|
+| Role: Marketing | Marketing-focused tour | Templates de marketing, dashboards de campanha |
+| Role: Developer | Technical tour | API docs, integracoes, CLI |
+| Role: Manager | Overview tour | Dashboards, reports, team features |
+| Goal: Save time | Efficiency path | Automations, templates, shortcuts |
+| Goal: Grow revenue | Growth path | Analytics, A/B testing, funnels |
+| Team: Solo | Individual setup | Personal workspace |
+| Team: 2-10 | Team setup | Invite flow, shared spaces |
+| Team: 10+ | Enterprise setup | Admin panel, roles, SSO |
+
+### 5. Empty State Strategy
+| Tela | Empty State Atual | Otimizado |
+|------|-------------------|----------|
+| Dashboard | "No data yet" | Pre-populated sample dashboard + "Start tracking" CTA |
+| Projects | "No projects" | Template gallery + "Create from template" CTA |
+| Team | "No members" | "Invite your team" + social proof ("Teams of 3+ are 2x more productive") |
+| Reports | "No reports" | Sample report + "Generate your first report" |
+
+### 6. Onboarding Emails (Lifecycle Sequence)
+| Email # | Timing | Subject | Conteudo | CTA |
+|---------|--------|---------|---------|-----|
+| 1 | Immediate | Welcome to [Product]! | Quick start guide (3 steps) | "Complete setup" |
+| 2 | D1 | Your first [action] in 2 min | Tutorial for core action | "Try now" |
+| 3 | D3 | Did you know about [feature]? | Secondary feature intro | "Explore" |
+| 4 | D5 | [Name], your progress | Checklist progress summary | "Keep going" |
+| 5 | D7 | Teams get 2x more value | Team invite prompt | "Invite team" |
+| 6 | D14 | Your first 2 weeks recap | Value delivered summary | "Upgrade" or "Share" |
+
+### 7. Onboarding Metrics
+| Metrica | Formula | Target |
+|---------|---------|--------|
+| Signup-to-Setup | Completed setup / Started signup | > 60% |
+| Setup Completion Time | Median time to finish onboarding | < 3 min |
+| Checklist Completion | Items completed / Total items | > 60% |
+| Quick Win Rate | Users reaching first value / Signups | > 50% |
+| D1 Return Rate | Users returning D1 / Signups | > 30% |
+| Onboarding NPS | NPS survey after onboarding | > 40 |
+| Skip Rate | Users who skip onboarding / Total | < 30% |
+| Feature Discovery Rate | Users using 3+ features D7 / Total | > 40% |
+
+## Saida
+- Onboarding flow design (visual)
+- Personalization matrix
+- Empty state designs
+- Email lifecycle sequence
+- Checklist specification
+- Metrics dashboard
+
+## Validacao
+- [ ] Pattern de onboarding selecionado com justificativa
+- [ ] Flow de 5 fases documentado
+- [ ] Personalizacao com pelo menos 3 branches
+- [ ] Empty states otimizados
+- [ ] Email sequence (6 emails, D0-D14)
+- [ ] Checklist com max 5 items
+- [ ] Metricas de onboarding definidas

package/squad-growth/tasks/design-reengagement-campaigns.md

@@ -0,0 +1,147 @@
+---
+task: design-reengagement-campaigns
+responsavel: "@ga-growth-hacker"
+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"
+---
+
+# Task: Design Re-engagement Campaigns
+
+## Metadata
+- **Squad:** squad-growth
+- **Agent:** Lever (ga-growth-hacker)
+- **Complexity:** Standard
+
+## Objetivo
+Desenhar campanhas de re-engajamento — criar estrategia multi-canal para reativar usuarios inativos e churned, segmentando por tempo de inatividade, valor e probabilidade de retorno.
+
+## Entrada
+- Inactive/churned user segments
+- Last activity data
+- Historical value (LTV/spend)
+- Previous engagement patterns
+- Churn reason data (if available)
+
+## Passos
+
+### 1. Inactivity Segmentation
+| Segmento | Inatividade | Probabilidade de Retorno | Investimento |
+|---------|-----------|------------------------|-------------|
+| At Risk | 7-14 days | Alta (60-80%) | Alto — prevent churn |
+| Lapsed | 15-30 days | Media (30-50%) | Medio |
+| Dormant | 31-90 days | Baixa (10-25%) | Seletivo por valor |
+| Churned | 91-180 days | Muito baixa (5-10%) | Apenas high-value |
+| Lost | 180+ days | Minimal (< 5%) | Ultimo esforco ou sunset |
+
+### 2. Re-engagement Campaign Matrix
+| Segmento | Canal 1 | Canal 2 | Canal 3 | Incentivo |
+|---------|---------|---------|---------|----------|
+| At Risk | In-app message | Push notification | Email | Feature highlight |
+| Lapsed | Email sequence | Push | Retargeting ads | Discount/offer |
+| Dormant | Email win-back | Retargeting | Direct mail (high value) | Special offer |
+| Churned | Personal email/call | Retargeting | Social | Major upgrade/discount |
+| Lost | Final email | — | — | Survey + final offer |
+
+### 3. Email Sequence — At Risk (7-14 days)
+| Email | Timing | Subject Line | Conteudo | CTA |
+|-------|--------|-------------|---------|-----|
+| 1 | Day 7 | "We miss you, [Name]!" | Show what they're missing | "See what's new" |
+| 2 | Day 10 | "[X] happened while you were away" | Personalized updates/activity | "Catch up now" |
+| 3 | Day 14 | "Quick question, [Name]" | Ask if they need help | "Reply" or "Get help" |
+
+### 4. Email Sequence — Lapsed (15-30 days)
+| Email | Timing | Subject Line | Conteudo | CTA |
+|-------|--------|-------------|---------|-----|
+| 1 | Day 15 | "Things have changed since you left" | New features, improvements | "See what's new" |
+| 2 | Day 20 | "[Name], here's a gift for you" | Exclusive discount or credit | "Claim your offer" |
+| 3 | Day 25 | "Your [data/project] is waiting" | Remind of stored value | "Continue where you left off" |
+| 4 | Day 30 | "Last chance: [offer] expires tomorrow" | Urgency + final offer | "Come back now" |
+
+### 5. Email Sequence — Dormant/Churned (31-90+ days)
+| Email | Timing | Subject Line | Conteudo | CTA |
+|-------|--------|-------------|---------|-----|
+| 1 | Day 45 | "A lot has changed at [Product]" | Major updates roundup | "See the new [Product]" |
+| 2 | Day 60 | "[Name], we'd love you back" | Emotional + special offer | "Reactivate with [discount]" |
+| 3 | Day 90 | "Before we say goodbye..." | Ask why they left + final offer | "Tell us why" or "Come back" |
+
+### 6. Multi-Channel Orchestration
+```
+Day 0: User becomes "At Risk" (no activity 7 days)
+  │
+  ├─ Day 7: Email 1 (content-focused)
+  ├─ Day 8: Push notification (if enabled)
+  │
+  ├─ Day 10: Email 2 (personalized)
+  │    └─ If opened but no action: In-app banner next visit
+  │
+  ├─ Day 14: Email 3 (help-focused)
+  │    └─ If no open: Start retargeting ads
+  │
+  ├─ Day 20: Email 4 (incentive)
+  │    └─ Retargeting ads with offer
+  │
+  ├─ Day 30: Email 5 (urgency)
+  │    └─ If high-value: Personal outreach (CS team)
+  │
+  ├─ Day 60: Win-back email (major changes)
+  │
+  └─ Day 90: Sunset email (final + survey)
+       └─ If no response: Move to suppression list
+```
+
+### 7. Personalization Strategies
+| Tipo | Dados Usados | Exemplo |
+|------|-------------|---------|
+| Content-based | Last features used | "Your dashboard has new insights" |
+| Achievement-based | Progress/level | "You were at Level 4 — pick up where you left off" |
+| Social-based | Team/friends activity | "Your team posted 5 updates" |
+| Value-based | Revenue/savings generated | "You saved R$2,400 last quarter" |
+| Fear of loss | Data/content at risk | "Your projects will be archived in 30 days" |
+| Surprise | Random reward | "We saved you a gift — claim in 24h" |
+
+### 8. Re-engagement Metrics
+| Metrica | Formula | Target |
+|---------|---------|--------|
+| Reactivation rate | Reactivated users / Campaign recipients | > 10% (at-risk), > 5% (dormant) |
+| Email open rate | Opens / Delivered | > 25% |
+| Email click rate | Clicks / Opens | > 5% |
+| Return visit rate | Users who visited / Recipients | > 15% |
+| Re-retained D30 | Still active 30 days after reactivation | > 40% |
+| Revenue from reactivated | Revenue from segment / Total revenue | Track MoM |
+| Campaign ROI | (Revenue gained - Campaign cost) / Cost | > 3x |
+| Unsubscribe rate | Unsubs / Delivered | < 1% |
+
+## Saida
+- Re-engagement campaign strategy
+- Email sequences per segment
+- Multi-channel orchestration flow
+- Personalization playbook
+- Metrics dashboard spec
+- Suppression/sunset policy
+
+## Validacao
+- [ ] Segmentos de inatividade definidos (5 niveis)
+- [ ] Campanhas desenhadas por segmento
+- [ ] Email sequences com timing e copy
+- [ ] Multi-channel orchestration documentado
+- [ ] Personalizacao implementada
+- [ ] Sunset policy definida
+- [ ] Metricas de eficacia trackadas

package/squad-growth/tasks/design-referral-system.md

@@ -0,0 +1,169 @@
+---
+task: design-referral-system
+responsavel: "@ga-growth-hacker"
+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"
+---
+
+# Task: Design Referral System
+
+## Metadata
+- **Squad:** squad-growth
+- **Agent:** Lever (ga-growth-hacker)
+- **Complexity:** Advanced
+
+## Objetivo
+Desenhar sistema de referral — criar mecanismo de indicacao que transforma usuarios satisfeitos em canal de aquisicao, maximizando viral coefficient e minimizando custo por aquisicao.
+
+## Entrada
+- Product type and value proposition
+- Current user base size
+- Customer satisfaction data (NPS/CSAT)
+- Budget for referral incentives
+
+## Passos
+
+### 1. Referral Model Selection
+| Modelo | Mecanismo | Melhor Para | Exemplo |
+|--------|----------|------------|---------|
+| Single-sided | Reward only referrer | Low-cost products | Uber (credits) |
+| Double-sided | Reward both referrer + referred | Subscription/SaaS | Dropbox (storage) |
+| Tiered | Increasing rewards per referral | High-engagement products | Ambassador programs |
+| Community | Status/recognition, not money | Communities | Stack Overflow |
+| Affiliate | Revenue share | B2B, high-ticket | SaaS affiliate programs |
+
+### 2. Incentive Design
+| Tipo de Incentivo | Exemplos | Custo | Eficacia |
+|------------------|---------|-------|---------|
+| Credit/discount | R$50 off, 20% discount | Variable | Alta para e-commerce |
+| Extended service | +1 month free, extra storage | Low marginal | Alta para SaaS |
+| Cash/gift card | R$25 pix, gift card | Fixed | Alta mas menos sticky |
+| Premium features | Unlock features temporariamente | Zero marginal | Media |
+| Physical reward | Swag, merchandise | Fixed + shipping | Baixa eficacia, alta brand |
+| Charity donation | Donate per referral | Fixed | Boa para brand values |
+
+### 3. Viral Coefficient Calculation
+```
+K-Factor = i × c
+
+Where:
+  i = average invitations sent per user
+  c = conversion rate of invitations
+
+K > 1 = viral growth (each user brings > 1 new user)
+K = 0.5-1.0 = viral amplification (boosts other channels)
+K < 0.5 = minimal viral effect
+```
+
+| Componente | Otimizacao |
+|-----------|-----------|
+| Invitations (i) | Make sharing easy, multiple channels, prompt at moment of delight |
+| Conversion (c) | Strong landing page, social proof, clear value prop, low friction signup |
+| Cycle time | Reduce time between signup → value → share |
+
+### 4. Referral Flow Design
+```
+┌─────────────────────────────────────────┐
+│ 1. TRIGGER                              │
+│    User reaches "moment of delight"     │
+│    (e.g., first success, milestone)     │
+│              │                          │
+│              ▼                          │
+│ 2. PROMPT                               │
+│    In-app prompt, email, or push        │
+│    "Share with friends, get [reward]"   │
+│              │                          │
+│              ▼                          │
+│ 3. SHARE                                │
+│    WhatsApp, Email, Link, Social        │
+│    Unique referral code/link            │
+│              │                          │
+│              ▼                          │
+│ 4. LANDING                              │
+│    Personalized landing page            │
+│    "[Friend] invited you to [Product]"  │
+│    Social proof + reward display        │
+│              │                          │
+│              ▼                          │
+│ 5. SIGNUP + ACTIVATION                  │
+│    Streamlined onboarding               │
+│    Referral code auto-applied           │
+│              │                          │
+│              ▼                          │
+│ 6. REWARD                               │
+│    Both parties rewarded                │
+│    Referrer notified of success         │
+│    Prompt referrer to share again       │
+│              │                          │
+│              ▼                          │
+│ 7. LOOP RESTART                         │
+│    New user reaches delight → shares    │
+└─────────────────────────────────────────┘
+```
+
+### 5. Sharing Channels
+| Canal | Conversao Tipica | UX | Implementacao |
+|-------|-----------------|-----|-------------|
+| WhatsApp | Alta (BR market) | Deep link + preview | WhatsApp Share API |
+| Email | Media | Personalized invite | Email API + template |
+| Copy link | Media | Simple, universal | Clipboard API |
+| SMS | Alta | Direct, personal | SMS API |
+| Social (Instagram, X) | Baixa | Broad reach | Social share APIs |
+| QR Code | Media | In-person | QR generator |
+
+### 6. Anti-Fraud Measures
+| Risco | Prevencao | Deteccao |
+|-------|----------|---------|
+| Self-referral | Different email, device fingerprint | IP/device matching |
+| Fake accounts | Email/phone verification | Cluster analysis |
+| Referral farms | Limit referrals per user/period | Velocity checks |
+| Coupon abuse | Single-use codes, linked to account | Code tracking |
+| Bot signups | CAPTCHA, behavior analysis | Bot detection |
+
+### 7. Referral Metrics Dashboard
+| Metrica | Formula | Target |
+|---------|---------|--------|
+| K-Factor | Invites sent × Invite conversion | > 0.5 |
+| Referral share of signups | Referred signups / Total signups | > 20% |
+| Invite rate | Users who shared / Total eligible | > 15% |
+| Invites per referrer | Total invites / Referrers | > 3 |
+| Invite conversion | Signups / Invites sent | > 10% |
+| Referred user LTV | LTV of referred vs organic | >= 1.0x |
+| CAC via referral | Total rewards / Referred signups | < other CAC |
+| Time to first referral | Days from signup to first share | < 14 days |
+| Referral program ROI | (Revenue from referred - Rewards) / Rewards | > 3x |
+
+## Saida
+- Referral system design document
+- Incentive model recommendation
+- Flow design with wireframes
+- Anti-fraud strategy
+- KPI dashboard spec
+- Implementation roadmap
+
+## Validacao
+- [ ] Modelo de referral selecionado com justificativa
+- [ ] Incentivos desenhados (both sides)
+- [ ] Flow de referral documentado (7 steps)
+- [ ] Canais de share priorizados
+- [ ] Anti-fraud measures definidas
+- [ ] Metricas de tracking configuradas
+- [ ] K-Factor target definido

package/squad-growth/tasks/design-viral-loop.md

@@ -0,0 +1,104 @@
+---
+task: design-viral-loop
+responsavel: "@ga-growth-hacker"
+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"
+---
+
+# Task: Design Viral Loop
+
+## Metadata
+- **Squad:** squad-growth
+- **Agent:** Lever (ga-growth-hacker)
+- **Complexity:** Advanced
+
+## Objetivo
+Design viral growth mechanisms — calculate k-factor, optimize viral cycle time, design incentive structures, select sharing mechanics and channels, and build a measurement framework for viral growth.
+
+## Entrada
+- Product/service model
+- Current user base metrics
+- Sharing/invite capabilities
+- Target audience profile
+
+## Passos
+
+### 1. K-Factor Calculation
+| Component | Formula | Target |
+|-----------|---------|--------|
+| K-factor | k = invites_per_user x conversion_rate | k > 1 = viral |
+| Invites per user | Total invites sent / Total users | Maximize |
+| Invite conversion | Signups from invites / Total invites | Maximize |
+| Viral cycle time | Average time from signup to first invite sent | Minimize |
+
+### 2. Viral Cycle Time Optimization
+- Map each step from user activation to invite sent
+- Identify friction points that delay sharing
+- Reduce steps between value moment and share prompt
+- A/B test timing of share prompts (immediate vs delayed)
+- Optimize for mobile-first sharing flows
+
+### 3. Incentive Design
+| Incentive Type | Description | Best For |
+|---------------|-------------|----------|
+| Double-sided | Both referrer and referee get reward | Marketplaces, SaaS |
+| One-sided (referrer) | Only referrer rewarded | Low-cost products |
+| One-sided (referee) | Only new user rewarded | High-value products |
+| Tiered | Rewards increase with more referrals | Power users |
+| Social proof | Status, badges, leaderboards | Community products |
+
+### 4. Sharing Mechanics
+- Direct invite (email, SMS, messaging apps)
+- Social sharing (public posts, stories)
+- Content sharing (user-generated, achievements)
+- Collaborative features (shared workspaces, multiplayer)
+- Embed/widget distribution
+
+### 5. Channel Selection
+| Channel | Friction | Reach | Conversion | Best For |
+|---------|---------|-------|-----------|----------|
+| WhatsApp/SMS | Low | Personal | High | 1:1 products |
+| Email | Medium | Professional | Medium | B2B, SaaS |
+| Social media | Low | Broad | Low | Consumer apps |
+| In-product | None | Existing users | High | Collaboration tools |
+
+### 6. Measurement Framework
+- K-factor by channel and cohort
+- Viral cycle time (days)
+- Invite send rate (% of users who invite)
+- Invite accept rate (% of invites that convert)
+- Viral attribution (% of new users from viral)
+- Cost per viral acquisition vs paid
+
+## Saida
+- Viral loop design document
+- K-factor model with projections
+- Incentive structure specification
+- Channel strategy and prioritization
+- Measurement dashboard spec
+
+## Validacao
+- [ ] K-factor formula defined with current baseline
+- [ ] Viral cycle time mapped with optimization opportunities
+- [ ] Incentive structure designed with cost modeling
+- [ ] Sharing mechanics selected and prioritized
+- [ ] Measurement framework with tracking events defined
+- [ ] A/B test plan for viral optimizations

package/squad-growth/tasks/formulate-test-hypotheses.md

@@ -0,0 +1,110 @@
+---
+task: formulate-test-hypotheses
+responsavel: "@ga-cro-specialist"
+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"
+---
+
+# Task: Formulate Test Hypotheses
+
+## Metadata
+- **Squad:** squad-growth
+- **Agent:** Convert (ga-cro-specialist)
+- **Complexity:** Standard
+
+## Objetivo
+Formular hipoteses de teste estruturadas — transformar insights e friction points em hipoteses testaveise priorizadas por ICE/PIE scoring.
+
+## Entrada
+- CRO audit findings
+- Friction points identified
+- Funnel analysis
+- Qualitative research insights
+
+## Passos
+
+### 1. Hypothesis Framework
+```
+IF we [change/add/remove this]
+THEN [this metric] will [increase/decrease]
+BECAUSE [data-supported reasoning]
+```
+
+**Exemplo:**
+```
+IF we simplify the signup form from 8 to 4 fields
+THEN the form completion rate will increase by 20%
+BECAUSE heatmap data shows 45% of users abandon after field 5,
+and industry benchmarks show 4-field forms convert 25% higher
+```
+
+### 2. Hypothesis Template
+| Campo | Descricao |
+|-------|----------|
+| ID | H-001 |
+| Hypothesis | IF... THEN... BECAUSE... |
+| Primary metric | Metric to measure success |
+| Secondary metrics | Supporting metrics to watch |
+| Guardrail metrics | Metrics that should NOT degrade |
+| Evidence | Data supporting the hypothesis |
+| Variant description | What changes in the variant |
+| Target page/flow | Where the test runs |
+| ICE Score | Impact × Confidence × Ease |
+
+### 3. ICE Scoring Framework
+| Dimensao | 1-3 (Low) | 4-7 (Medium) | 8-10 (High) |
+|----------|----------|-------------|------------|
+| Impact | Minor metric move | Moderate improvement | Major conversion lift |
+| Confidence | Gut feeling | Some data support | Strong data + precedent |
+| Ease | Complex, multi-team | Moderate effort | Quick, single change |
+
+### 4. Hypothesis Backlog (Prioritized)
+| ID | Hypothesis (short) | Impact | Confidence | Ease | ICE | Priority |
+|----|-------------------|--------|-----------|------|-----|---------|
+| H-001 | Simplify signup form | 8 | 7 | 9 | 504 | P1 |
+| H-002 | Add social proof to LP | 6 | 8 | 8 | 384 | P2 |
+| H-003 | Redesign pricing page | 9 | 5 | 4 | 180 | P3 |
+| H-004 | | | | | | |
+
+### 5. Hypothesis Quality Checklist
+| Criterio | Check |
+|----------|-------|
+| Data-supported | Evidence cited, not opinion |
+| Specific metric | Clear primary metric defined |
+| Measurable | Can calculate lift and significance |
+| Testable | Can create meaningful variant |
+| Time-bound | Reasonable test duration estimated |
+| No multiple variables | Tests ONE change at a time |
+| Guardrails defined | Won't harm other key metrics |
+
+## Saida
+- Hypothesis backlog (prioritized)
+- ICE scores for each hypothesis
+- Evidence documentation per hypothesis
+- Hypothesis quality validation
+
+## Validacao
+- [ ] Hipoteses seguem formato IF/THEN/BECAUSE
+- [ ] Cada hipotese tem evidence documental
+- [ ] ICE score calculado
+- [ ] Backlog priorizado por ICE
+- [ ] Guardrail metrics definidos
+- [ ] Quality checklist passado para top 5