iron-edge 1.0.0

package/squads/squad-growth/knowledge-base/plg-metrics-framework.md

@@ -0,0 +1,183 @@
+# Knowledge Base: Product-Led Growth (PLG) Metrics Framework
+
+> Fonte principal: Wes Bush "Product-Led Growth" (2019), OpenView Partners, Elena Verna, Kyle Poyar
+
+## O Que e PLG
+
+Product-Led Growth e uma estrategia go-to-market onde o proprio produto e o principal veiculo de aquisicao, conversao e expansao. O termo foi popularizado pela OpenView Partners em 2016 e sistematizado por Wes Bush.
+
+**Principios fundamentais:**
+1. O produto e o canal — usuario experimenta valor antes de falar com vendas
+2. Self-serve first — onboarding sem fricao, sem demo obrigatoria
+3. Time-to-value minimo — usuario atinge o "aha moment" o mais rapido possivel
+4. Bottom-up adoption — usuarios individuais adotam, depois a organizacao compra
+5. Data-driven expansion — upsell baseado em comportamento, nao em pitch
+
+**Exemplos canonicos:**
+| Empresa | Modelo PLG | Mecanismo |
+|---------|-----------|-----------|
+| Slack | Freemium + viral | Teams adotam, depois compram |
+| Zoom | Freemium + viral | Anfitriao usa, convidados experimentam |
+| Figma | Freemium + colaboracao | Designers compartilham, times adotam |
+| Notion | Freemium + templates | Templates atraem, workflows retêm |
+| Calendly | Freemium + viral | Cada agendamento expoe a marca |
+| Canva | Freemium + content | Designs compartilhados atraem |
+| Loom | Freemium + viral | Cada video enviado e marketing |
+
+---
+
+## Modelos de Monetizacao PLG
+
+### Freemium
+- Plano gratuito permanente com limitacoes
+- Conversao tipica: 2-5% para planos pagos (faixa "great": 5%+)
+- Ideal para: produtos com alto volume e low touch
+- Risco: usuarios "free forever" que consomem recursos sem converter
+
+### Free Trial (time-limited)
+- Acesso completo por periodo limitado (7, 14, 30 dias)
+- Conversao tipica: 8-12% (faixa "good"), 15-25% (faixa "great")
+- Ideal para: produtos complexos que precisam de tempo para demonstrar valor
+- Risco: pressao temporal pode frustrar usuarios
+
+### Reverse Trial (modelo emergente)
+- Usuario comeca com funcionalidades premium, apos periodo faz downgrade para free
+- Conversao tipica: 10-15%
+- Ideal para: produtos onde o valor premium e claro mas nao imediato
+- Usado por: Ahrefs (acesso limitado gratuito com trial premium)
+
+### Hybrid PLG + Sales (PLG + SLG)
+- Self-serve para SMBs, sales-assisted para Enterprise
+- Usado por: Slack, Notion, Figma, Datadog
+- PQLs (Product Qualified Leads) alimentam o time de vendas
+
+---
+
+## Product Qualified Leads (PQLs)
+
+Usuarios que demonstraram intencao de compra atraves de comportamento no produto. Conceito sistematizado pela OpenView Partners.
+
+**Sinais tipicos de PQL:**
+- Atingiu limites do plano free
+- Convidou X colegas para o workspace
+- Usou feature premium durante trial
+- Excedeu volume de uso
+- Visitou pagina de pricing multiplas vezes
+- Exportou dados (sinal de que o dado e valioso)
+
+**PQL Scoring Model:**
+| Sinal | Peso | Threshold |
+|-------|------|-----------|
+| Usuarios no workspace | 5 | >= 5 |
+| Features premium usadas | 4 | >= 3 |
+| Visitas a pricing page | 3 | >= 2 em 7 dias |
+| Tempo ativo semanal | 3 | >= 3 horas |
+| Integracoes ativas | 2 | >= 2 |
+
+**PQL vs MQL:**
+| Aspecto | MQL (Marketing Qualified Lead) | PQL (Product Qualified Lead) |
+|---------|-------------------------------|------------------------------|
+| Origem do sinal | Comportamento em marketing (content, form) | Comportamento no produto |
+| Intencao | Interesse declarado | Intencao demonstrada por uso |
+| Predicao de conversao | Baixa | Alta |
+| Ciclo de venda | Mais longo | Mais curto |
+
+---
+
+## Onboarding como Alavanca de Growth
+
+Samuel Hulick (UserOnboard.com) demonstrou que a maioria dos produtos perde 40-60% dos usuarios no primeiro uso.
+
+### Framework de Onboarding
+1. **Sign-up Flow** — Minimo de campos. Social login. Sem cartao de credito (para freemium).
+2. **Welcome Survey** — 2-3 perguntas para personalizar experiencia (JTBD, role, objetivo).
+3. **Setup Checklist** — Passos claros com progresso visual.
+4. **Quick Win** — Levar ao primeiro valor em <5 minutos.
+5. **Celebrate** — Reforco positivo ao atingir marcos.
+6. **Ongoing Education** — Tooltips contextuais, emails educacionais.
+
+### Metricas de Onboarding
+| Metrica | Descricao | Benchmark |
+|---------|-----------|-----------|
+| **Time-to-First-Value (TTFV)** | Tempo ate o usuario atingir o primeiro valor | <5 min (ideal para PLG) |
+| **Activation Rate** | % que completa setup critico | >50% e considerado bom |
+| **D1 Retention** | % que volta no dia 1 | >40% indica boa ativacao |
+| **D7 Retention** | % que volta na semana 1 | >20% para apps, >50% SaaS |
+| **D30 Retention** | % que volta no mes 1 | >10% apps, >30% SaaS |
+| **Setup Completion Rate** | % que completa todo o onboarding | Meta: >70% |
+| **Onboarding drop-off step** | Onde mais usuarios abandonam | Identificar e otimizar |
+
+---
+
+## Metricas Fundamentais PLG
+
+### Activation
+```
+Activation Rate = Usuarios que atingiram "aha moment" / Total de novos usuarios × 100
+
+"Aha Moment" = primeiro momento em que o usuario experimenta o valor core do produto
+```
+
+**Definindo o aha moment:**
+- Analise de cohort: quais usuarios D30 ativos completaram qual acao no D1?
+- Correlacionar acoes D1 com retencao futura
+- Ex: Facebook "7 amigos em 10 dias" (Chamath Palihapitiya)
+
+### Retention (Curva de Retencao)
+```
+D1 Retention = Usuarios que voltaram no dia 1 / Usuarios que fizeram signup
+D7 Retention = Usuarios que voltaram na semana 1 / Usuarios que fizeram signup
+D30 Retention = Usuarios que voltaram no mes 1 / Usuarios que fizeram signup
+
+Cohort Retention Rate = Usuarios do cohort ativos no periodo N / Tamanho original do cohort × 100
+```
+
+**A curva de retencao e o grafico mais importante em growth:**
+- Curva que estabiliza = product-market fit
+- Curva que vai a zero = problema estrutural no produto
+
+### Expansion (NRR/NDR)
+```
+Net Revenue Retention (NRR) = (MRR inicio - churned MRR + expansion MRR) / MRR inicio × 100
+
+NRR > 100% = crescimento sem aquisicao nova (expansion supera churn)
+NRR benchmark: >100% saudavel, >120% excelente, >130% world-class
+```
+
+---
+
+## Network Effects em PLG (Andrew Chen, "The Cold Start Problem")
+
+**Lei de Metcalfe:** Valor da rede proporcional a n². Cada novo usuario adiciona valor para todos.
+
+**Lei de Reed:** Para redes com grupos, valor cresce exponencialmente (2^n). Cada usuario multiplica possibilidades de subgrupos (WhatsApp groups, Slack channels).
+
+**Tipos de efeitos de rede:**
+| Tipo | Descricao | Exemplo |
+|------|-----------|---------|
+| **Direto** | Mais usuarios = mais valor para cada usuario | WhatsApp, telefone |
+| **Indireto (cross-side)** | Mais usuarios de um lado = mais valor para outro | Uber (riders/drivers) |
+| **Data network effects** | Mais uso = melhor produto (via dados) | Waze, Google Search |
+| **Marketplace** | Liquidity atrai ambos os lados | Airbnb, Amazon |
+
+**Cold Start Problem:** Como comecar uma rede sem usuarios? Estrategias:
+1. **Small networks first** — Dominar um nicho pequeno antes de escalar (Uber: cidade por cidade)
+2. **Single-player mode** — Produto util mesmo sem outros usuarios (Notion, Dropbox)
+3. **Marquee users** — Trazer usuarios de alto valor que atraem outros
+4. **Manual outreach** — Crescimento manual ate atingir massa critica
+
+---
+
+## PLG Benchmarks (OpenView Partners, Lenny Rachitsky — 2025)
+
+| Metrica | SaaS B2B | SaaS B2C |
+|---------|----------|----------|
+| Freemium conversion (free → paid) | 2-5% | 1-3% |
+| Trial conversion (trial → paid) | 8-25% | 10-30% |
+| D1 Retention | 40-60% | 25-45% |
+| D7 Retention | 20-35% | 10-25% |
+| D30 Retention | 10-25% | 5-15% |
+| NRR (saudavel) | >100% | >90% |
+| NRR (excelente) | >120% | >110% |
+| Activation Rate (bom) | >50% | >40% |
+| Time-to-first-value (ideal PLG) | <5 min | <2 min |

package/squads/squad-growth/knowledge-base/privacy-compliance-analytics.md

@@ -0,0 +1,195 @@
+# Knowledge Base: Privacy & Compliance for Analytics
+
+## Regulatory Landscape
+
+### LGPD (Brazil)
+| Aspecto | Requisito |
+|---------|----------|
+| Scope | All data processing of Brazilian residents |
+| Legal bases | Consent, legitimate interest, contract, legal obligation |
+| Consent | Must be free, informed, unambiguous, specific |
+| Data subject rights | Access, correction, deletion, portability, objection |
+| DPO | Required (Encarregado de Dados) |
+| Penalties | Up to 2% of revenue (max R$50M per violation) |
+| Authority | ANPD (Autoridade Nacional de Protecao de Dados) |
+
+### GDPR (EU)
+| Aspecto | Requisito |
+|---------|----------|
+| Scope | All data processing of EU residents |
+| Consent | Prior, explicit, granular, withdrawable |
+| Cookie consent | Required before any non-essential cookies |
+| Penalties | Up to 4% of global revenue or EUR 20M |
+| Extra-territorial | Applies even if company is outside EU |
+
+### Key Differences
+| Area | LGPD | GDPR |
+|------|------|------|
+| Cookie consent | Less strict (legitimate interest possible) | Strict (opt-in required) |
+| DPO requirement | Always required | Only in specific cases |
+| Penalties | Lower (R$50M cap) | Higher (no cap on %) |
+| Cross-border transfer | Allowed with safeguards | Strict adequacy requirements |
+
+## Google Consent Mode v2
+
+### Consent States
+| Signal | Controls | Default (no consent) | Granted |
+|--------|---------|---------------------|---------|
+| ad_storage | Ad cookies (Google Ads, Floodlight) | Denied | Cookies set |
+| ad_user_data | User data sent to Google for ads | Denied | Data shared |
+| ad_personalization | Personalized advertising | Denied | Ads personalized |
+| analytics_storage | Analytics cookies (GA4) | Denied | Full tracking |
+| functionality_storage | Essential functionality | Granted | Granted |
+| personalization_storage | Personalization (recommendations) | Denied | Personalized |
+| security_storage | Security (fraud detection) | Granted | Granted |
+
+### Implementation
+```javascript
+// Default consent state (before user interaction)
+gtag('consent', 'default', {
+  ad_storage: 'denied',
+  ad_user_data: 'denied',
+  ad_personalization: 'denied',
+  analytics_storage: 'denied',
+  functionality_storage: 'granted',
+  security_storage: 'granted',
+  wait_for_update: 500  // ms to wait for CMP
+});
+
+// After user grants consent
+gtag('consent', 'update', {
+  ad_storage: 'granted',
+  ad_user_data: 'granted',
+  ad_personalization: 'granted',
+  analytics_storage: 'granted'
+});
+```
+
+### Consent Mode v2 Behavior
+| Consent State | GA4 Behavior | Google Ads Behavior |
+|-------------|-------------|-------------------|
+| analytics_storage: denied | Cookieless pings (modeled data) | — |
+| ad_storage: denied | — | No conversion cookies, modeled |
+| Both granted | Full measurement | Full conversion tracking |
+
+## Consent Management Platform (CMP)
+
+### Requirements
+| Requirement | Implementation |
+|------------|---------------|
+| Banner visibility | Show on first visit, every page |
+| Granular choices | Per-category consent (not just accept all) |
+| Easy rejection | Reject must be as easy as accept |
+| Persistent storage | Remember choice for 6-12 months |
+| Re-consent | Allow users to change preferences anytime |
+| Documentation | Log consent with timestamp, version, choices |
+
+### CMP Tools
+| Tool | Type | LGPD | GDPR | Google CMP Partner |
+|------|------|------|------|-------------------|
+| Cookiebot | Cloud | Yes | Yes | Yes |
+| OneTrust | Enterprise | Yes | Yes | Yes |
+| Osano | Cloud | Yes | Yes | Yes |
+| Cookie Script | Cloud | Yes | Yes | Yes |
+| Custom | Self-built | Configurable | Configurable | Manual integration |
+
+## Data Collection Principles
+
+### Minimization
+| Rule | Implementation |
+|------|---------------|
+| Collect only necessary data | Audit every event parameter |
+| No PII in analytics | Never send email, phone, name to GA4 |
+| Hash sensitive data | SHA-256 for enhanced conversions |
+| IP anonymization | GA4 does this by default |
+| Data retention limits | Set GA4 retention to 14 months max |
+| Delete on request | Process data deletion requests |
+
+### PII Checklist
+| Data | PII? | In GA4? | In Data Layer? |
+|------|------|---------|---------------|
+| Email address | Yes | NEVER | Hashed only (enhanced conv) |
+| Phone number | Yes | NEVER | Hashed only |
+| Full name | Yes | NEVER | NEVER |
+| IP address | Yes | Auto-anonymized | NEVER raw |
+| User ID | Pseudo-PII | Yes (as user_id) | Yes |
+| Device ID | Pseudo-PII | Auto-collected | No |
+| Credit card | Yes | NEVER | NEVER |
+| CPF/SSN | Yes | NEVER | NEVER |
+
+## Server-Side Tracking & Privacy
+
+### Benefits for Privacy
+| Benefit | How |
+|---------|-----|
+| Data control | Filter PII before sending to platforms |
+| First-party context | Cookies set from your domain |
+| Reduced client exposure | Less JavaScript, less data in browser |
+| Consent enforcement | Server validates consent before forwarding |
+| Ad blocker resilience | Server-to-server, no browser blocking |
+
+### Architecture
+```
+User Browser
+  → First-party cookie (your domain)
+  → GTM Client-side (minimal tags)
+  → Your Server (GTM Server-Side)
+    → Filter PII
+    → Check consent state
+    → Forward to GA4 (server-to-server)
+    → Forward to Meta CAPI (server-to-server)
+    → Forward to Google Ads (server-to-server)
+    → Log to Data Warehouse
+```
+
+## Audit & Compliance Checklist
+| Check | Frequency | Responsible |
+|-------|----------|-------------|
+| Consent banner functional | Monthly | Analytics team |
+| No PII in GA4 | Monthly | Analytics team |
+| Data retention settings correct | Quarterly | DPO |
+| Data deletion requests processed | As needed | DPO |
+| Privacy policy up-to-date | Quarterly | Legal |
+| CMP configuration reviewed | Quarterly | Analytics + Legal |
+| Server-side PII filtering | Monthly | Engineering |
+| Cookie audit (what's set) | Quarterly | Analytics team |
+| Third-party script audit | Quarterly | Engineering |
+| Consent logs archived | Continuous | CMP |
+
+## LGPD (Lei Geral de Protecao de Dados)
+
+### Key Requirements
+- 10 lawful bases (consent, legitimate interest, contract, etc.)
+- Data subject rights: access, correction, deletion, portability
+- DPO requirement for large-scale processing
+- Penalties: up to 2% of revenue (capped at R$50M per violation)
+
+## Consent Mode v2 (Google)
+
+### Overview
+- Required for EEA traffic since March 2024
+- **Basic mode:** no tags fire without consent, conversion modeling fills gaps
+- **Advanced mode:** cookieless pings sent, better modeling accuracy
+- **Implementation:** Google Tag Manager consent template
+
+## Server-Side Tracking
+
+### GTM Server-Side
+- Runs on cloud (GCP, AWS), first-party context
+- **Benefits:** bypasses ad blockers, first-party cookies, data control
+- **Setup:** custom domain, cloud container, tag migration
+
+## Data Anonymization
+
+### Techniques
+- IP masking (last octet zeroing)
+- User ID hashing (SHA-256 with salt)
+- k-anonymity (minimum group size of 5)
+- Aggregation thresholds for reporting
+
+## Privacy-First Architecture
+
+### Principles
+- **Data minimization:** collect only what's needed
+- **Purpose limitation:** use data only for stated purpose
+- **Cookieless approaches:** fingerprint-free cohorts, server-side enrichment, first-party data strategies

package/squads/squad-growth/knowledge-base/retention-lifecycle-marketing.md

@@ -0,0 +1,193 @@
+# Knowledge Base: Retention & Lifecycle Marketing
+
+> Fonte: Nir Eyal "Hooked" (2014), Casey Winters (Pinterest, Grubhub), Elena Verna, Reforge
+
+## Retencao como Fundamento do Growth
+
+Casey Winters: "Se voce nao consegue reter usuarios, voce nao tem um negocio — voce tem um vazamento."
+
+**A curva de retencao e o grafico mais importante em growth:**
+- Curva que estabiliza (flattens) = product-market fit
+- Curva que vai a zero = problema estrutural no produto
+
+**Tipos de retencao:**
+| Tipo | O que mede | Metrica |
+|------|-----------|---------|
+| **User retention** | O usuario retorna ao produto | DAU/MAU, D1/D7/D30 |
+| **Revenue retention** | A receita se mantem/expande | MRR retention, NRR |
+| **Engagement retention** | O nivel de uso se mantem | Sessions/user, features used |
+
+**Benchmarks de retencao (Lenny Rachitsky, 2025):**
+| Tipo de produto | D1 | D7 | D30 |
+|----------------|----|----|-----|
+| Consumer social/mobile | 25-45% | 10-25% | 5-15% |
+| SaaS B2B | 40-60% | 20-35% | 10-25% |
+| E-commerce | 30-50% | 15-30% | 8-20% |
+
+---
+
+## Hook Model (Nir Eyal — "Hooked", 2014)
+
+Como produtos criam habitos em 4 fases:
+
+```
+Trigger → Action → Variable Reward → Investment
+   ↑                                        |
+   └────────────────────────────────────────┘
+```
+
+### 1. Trigger (Gatilho)
+| Tipo | Descricao | Objetivo |
+|------|-----------|----------|
+| **Externo** | Push notification, email, ad, mencao social | Initiar o loop nos primeiros usos |
+| **Interno** | Tedio, solidao, incerteza, FOMO | Objetivo final — usuario se auto-aciona |
+
+Evolucao: Produto deve graduar o usuario de triggers externos para internos. Quando o usuario abre Instagram ao se sentir entediado (sem notificacao), o trigger interno foi criado.
+
+### 2. Action (Acao)
+- O comportamento mais simples em antecipacao de recompensa
+- BJ Fogg: B = MAT (Motivation × Ability × Trigger)
+- Deve ser extremamente facil
+- Exemplos: abrir app, scrollar feed, digitar busca
+
+### 3. Variable Reward (Recompensa Variavel)
+A variabilidade e crucial — recompensas previsiveis perdem efeito (Skinner, psicologia comportamental).
+
+| Tipo | Descricao | Exemplo |
+|------|-----------|---------|
+| **Tribe** | Reconhecimento social | Likes, comentarios, resposta de amigo |
+| **Hunt** | Busca por recursos/informacao | Scroll infinito, search results |
+| **Self** | Dominio pessoal, competencia | Gamificacao, progresso, conquistas |
+
+### 4. Investment (Investimento)
+O usuario investe algo que aumenta switching costs e probabilidade de retorno.
+- **Tempo:** Completar perfil, configurar workspace
+- **Dados:** Adicionar itens, customizar preferencias
+- **Conteudo:** Criar posts, uploads
+- **Reputacao:** Seguidores, reviews, credibilidade
+- **Social:** Adicionar amigos, conexoes
+
+---
+
+## Lifecycle Marketing
+
+### Fases do ciclo de vida e estrategias
+
+| Fase | Periodo | Objetivo | Canais | Exemplos de Acoes |
+|------|---------|----------|--------|------------------|
+| **Onboarding** | D0-D7 | Ativacao, primeiro valor | Email, in-app, push | Tutorial interativo, checklist |
+| **Activation** | D7-D30 | Habito, uso recorrente | Email, in-app | Tips de features, case studies |
+| **Growth** | D30-D90 | Expansao, upgrade | Email, in-app, sales | PQL triggers, trial de features premium |
+| **Maturity** | D90+ | Retencao, advocacy | Email, community | NPS, referral, conteudo avancado |
+| **Decline** | Queda em uso | Re-engagement | Email, push, retargeting | "Sentimos sua falta", novidades |
+| **Churn** | Inativo | Resurrection | Email, ads | Win-back offer, novo feature |
+
+### Drip Campaign Framework (Onboarding)
+
+```
+Dia 0:  Welcome + Quick Win (valor imediato)
+Dia 2:  Feature highlight #1 (core value)
+Dia 5:  Case study (social proof)
+Dia 8:  Feature highlight #2 (secondary value)
+Dia 12: Educational content (thought leadership)
+Dia 15: Upgrade offer / CTA principal
+Dia 20: Final follow-up (urgencia)
+```
+
+---
+
+## Churn Analysis
+
+### Tipos de churn
+| Tipo | Descricao | Como reduzir |
+|------|-----------|-------------|
+| **Voluntary churn** | Usuario decide sair | Melhorar produto, customer success |
+| **Involuntary churn** | Falha de pagamento, cartao expirado | Dunning emails, Smart Retry |
+| **Logo churn** | Clientes perdidos (headcount) | Monitorar sinal de alerta |
+| **Revenue churn** | Receita perdida | Pode ser diferente de logo churn |
+
+### Involuntary Churn — Dunning
+- Representa 20-40% do churn total em SaaS
+- **Dunning** = sequencia de emails para recuperar pagamentos falhados
+- Smart Retry: testar diferentes horarios/dias para reprocessar cartao
+- Alertas proativos antes de expirar
+
+### Metodos de Analise de Churn
+1. **Survival analysis** — Kaplan-Meier curves para estimar probabilidade de churn ao longo do tempo
+2. **Cohort analysis** — Retencao por cohort para identificar tendencias
+3. **Behavioral segmentation** — Quais comportamentos precedem churn?
+4. **Exit surveys** — Perguntar diretamente por que o usuario esta saindo
+5. **Predictive modeling** — ML para identificar usuarios em risco antes do churn
+
+### Sinais de churn (leading indicators)
+- Queda no login frequency (indicador mais forte)
+- Reducao no uso de features core
+- Tickets de suporte nao resolvidos
+- Nao-adocao de novas features
+- Reducao no numero de usuarios ativos na conta
+- Mudanca de champion interno (compra enterprise)
+
+### Churn Rate Formulas
+```
+Monthly Logo Churn = Clientes perdidos no mes / Clientes no inicio do mes × 100
+
+Monthly Revenue Churn = MRR perdido no mes / MRR no inicio do mes × 100
+
+Net Revenue Retention (NRR) = (MRR inicio - churn MRR + expansion MRR) / MRR inicio × 100
+
+Benchmarks NRR: >100% saudavel, >120% excelente, >130% world-class (Bessemer)
+```
+
+---
+
+## Resurrection Campaigns
+
+Para reconquistar usuarios que churned ou se tornaram inativos.
+
+### Abordagens por situacao
+| Abordagem | Quando usar | Mensagem |
+|-----------|-------------|---------|
+| **What's new** | Produto melhorou muito | "Desde que voce foi, adicionamos X, Y, Z" |
+| **Win-back offer** | Usuario estava satisfeito | Desconto temporario para retornar |
+| **Personalized value** | Usuario criou dados no produto | "Seus dados ainda estao aqui" |
+| **Social proof** | Produto cresceu muito | "150.000 novas empresas entraram" |
+| **New use case** | Usuario tinha caso de uso limitado | Apresentar novo caso de uso |
+
+### Timing de campanha de win-back
+```
+Dia 1 do churn:    Email transacional de confirmacao (se aplicavel)
+Dia 7:             "Sentimos sua falta" + novidades
+Dia 30:            Offer de desconto para retornar
+Dia 60:            Testemunho de customer relevante
+Dia 90:            Final email com oferta especial + ask for feedback
+Pos 90 dias:       Transferir para ads retargeting
+```
+
+---
+
+## Segmentacao para Lifecycle Marketing
+
+| Criterio | Exemplo | Acao |
+|----------|---------|------|
+| **Lifecycle stage** | "Trial expira em 3 dias" | Email urgente de upgrade |
+| **Comportamental** | "Usou feature X mas nao Y" | Tutorial sobre feature Y |
+| **Engagement** | "Abriu ultimos 5 emails" | Oferta exclusiva para engajados |
+| **RFM** | "Alto valor, baixa frequencia" | Re-engagement com desconto |
+| **Predictive** | "Alta probabilidade de churn" | Intervencao de customer success |
+| **Feature adoption** | "Nao usou collaboration features" | Educar sobre casos de uso |
+
+---
+
+## DAU/MAU Ratio — Stickiness Metric
+
+```
+DAU/MAU = Usuarios ativos diarios / Usuarios ativos mensais × 100
+
+Benchmarks:
+  > 50%: Produto com habito diario (Facebook, WhatsApp)
+  > 25%: Bom para SaaS B2B
+  > 10%: Aceitavel para SaaS de uso semanal
+  < 10%: Sinal de alerta de engajamento baixo
+```
+
+**Interpretacao:** DAU/MAU de 50% significa que o usuario medio usa o produto 15 dias por mes (50% × 30 dias). Para produtos de uso diario como comunicacao e social, target deve ser >50%.

package/squads/squad-growth/knowledge-base/routing-catalog.md

@@ -0,0 +1,65 @@
+# Routing Catalog — Growth Analytics
+
+Maps keywords and domains to specialist agents for intelligent routing within the Growth squad.
+
+## Domain Routing Matrix
+
+| Domain | Keywords | Primary Agent | Secondary Agent |
+|--------|----------|--------------|-----------------|
+| SEO strategy | SEO, search engine, organic traffic, rankings, keywords, backlinks, topical authority, programmatic SEO | ga-seo-strategist | ga-data-analyst |
+| Growth hacking | growth hacking, viral, referral, activation, network effects, flywheel, PLG, growth loop, K-factor | ga-growth-hacker | growth-orqx |
+| Analytics & data | analytics, GA4, dashboard, metrics, KPIs, data analysis, reporting, cohort, retention | ga-data-analyst | ga-analytics-engineer |
+| Analytics engineering | tracking, tag manager, GTM, data pipeline, event schema, data layer, BigQuery, dbt | ga-analytics-engineer | ga-data-analyst |
+| CRO (organic) | conversion rate, A/B test, experimentation, optimization, funnel, landing page, checkout | ga-cro-specialist | ga-growth-hacker |
+| Campaign analysis | campaign performance, channel attribution, ROI analysis, spend analysis, CAC, LTV | ga-campaign-analyst | ga-data-analyst |
+| Orchestration & strategy | growth strategy, OKRs, growth model, north star metric, unit economics, AARRR | growth-orqx | ga-growth-hacker |
+| Retention & lifecycle | retention, churn, lifecycle email, win-back, engagement, DAU/MAU, Hook Model | ga-growth-hacker | ga-data-analyst |
+| Referral & viral | referral program, viral coefficient, K-factor, referral mechanics, incentive design | ga-growth-hacker | ga-cro-specialist |
+| Product-led growth | PLG, PQL, onboarding, time-to-value, freemium, trial conversion, activation | ga-growth-hacker | ga-data-analyst |
+| Email marketing | email automation, drip campaign, deliverability, segmentation, ESP, lifecycle | ga-campaign-analyst | ga-growth-hacker |
+| Community-led growth | community, DevRel, CLG, SPACES, Discord, forum, developer relations | ga-growth-hacker | growth-orqx |
+| Brazilian market | PIX, LGPD, RD Station, Hotmart, infoprodutos, WhatsApp marketing, BR SEO | growth-orqx | ga-seo-strategist |
+| Attribution | multi-touch attribution, attribution model, incrementality, MMM, data-driven | ga-analytics-engineer | ga-campaign-analyst |
+| Experimentation | A/B test, Bayesian, frequentist, multi-armed bandit, Statsig, statistical significance | ga-cro-specialist | ga-analytics-engineer |
+
+## Task File Mapping
+
+| Mission Keyword | Task File | Agent |
+|----------------|-----------|-------|
+| `funil` | `analyze-conversion-funnel.md` | @ga-data-analyst |
+| `canais` | `analyze-channel-performance.md` | @ga-campaign-analyst |
+| `cac` | `analyze-cac-by-channel.md` | @ga-analytics-engineer |
+| `retencao` | `analyze-retention-churn.md` | @ga-data-analyst |
+| `kpi` | `build-kpi-tree-dashboard.md` | @ga-analytics-engineer |
+| `seo` | `audit-sitemap-robots.md` | @ga-seo-strategist |
+| `backlinks` | `analyze-backlink-profile.md` | @ga-seo-strategist |
+| `tracking` | `audit-tracking-accuracy.md` | @ga-analytics-engineer |
+| `experimento` | `analyze-experiment-results.md` | @ga-growth-hacker |
+| `referral` | `build-referral-program.md` | @ga-growth-hacker |
+| `cro` | `analyze-funnel-segmented.md` | @ga-cro-specialist |
+
+## Multi-Domain Requests — Routing Logic
+
+Quando um request abrange multiplos dominios, o growth-orqx coordena:
+
+```
+Request: "Analisar por que o funil de conversao caiu 20% esse mes"
+  → ga-data-analyst: Analise de funil (dados quantitativos)
+  → ga-cro-specialist: Hipoteses de conversao (qualitativo + experimentos)
+  → ga-campaign-analyst: Verificar mudancas em canais (attribution)
+  → growth-orqx: Sintetizar e recomendar plano de acao
+
+Request: "Construir estrategia de growth para novo produto SaaS"
+  → growth-orqx: Estrategia geral (AARRR, loops, NSM)
+  → ga-growth-hacker: PLG, viral, referral mechanics
+  → ga-seo-strategist: Canal SEO e content
+  → ga-data-analyst: Definir metricas e dashboards
+```
+
+## Usage Notes
+
+- Route to the **Primary Agent** by default for each domain.
+- Use the **Secondary Agent** when the request spans two domains.
+- For multi-domain requests, the orchestrator (growth-orqx) coordinates handoffs.
+- For requests com keyword "brasil" ou "BR" — sempre incluir contexto de `brazilian-growth-context.md`.
+- Para qualquer experimento, verificar `experimentation-boundaries.md` para ownership.