growth-enginelisted

# Growth Engine Autonomous experimentation framework that goes beyond basic A/B testing. Combines Bayesian inference, multi-armed bandits, CUPED variance reduction, and experiment dependency graphs to run statistically rigorous growth experiments across all marketing channels. ## When to Use - Run experiment on [channel] - Check for conflicting experiments - Score this test - What's in the playbook for [channel]? - Calculate sample size for [metric] with [MDE] - Switch to bandit mode ## Workflow 1. **Define Hypothesis** -- state expected lift, primary metric, guardrail metrics 2. **Check Dependency Graph** -- verify no conflicting experiments are running 3. **Calculate Sample Size** -- CUPED-adjusted MDE with channel-specific minimums 4. **Select Testing Mode** -- Bayesian (fixed horizon), Bandit (continuous), or Sequential (early stopping) 5. **Run Experiment** -- log daily variant metrics 6. **Score Results** -- posterior distributions + credible intervals (Bayesian) or confidence intervals (frequentist) 7. **Decision Gate** -- requires >90% posterior probability AND >10% relative lift to declare winner 8. **Archive to Playbook** -- winning variant, context, screenshots, follow-up ideas ## Reference Full methodology lives in [README.md](README.md) in this folder: configuration, scoring formulas, output formats, integration points, and worked examples. Read it before producing scored output.