Ab Test Analysis

This handles the full statistical analysis workflow for A/B tests: validates sample sizes and test duration, calculates significance with p-values and confidence intervals, checks for sample ratio mismatches, and gives you a clear ship/extend/stop recommendation. It can read CSV or Excel files directly and generates Python scripts for the calculations. The best part is the decision framework that accounts for guardrail metrics, so you won't ship a win that tanks revenue or engagement elsewhere. It walks through hypothesis validation, effect size calculations, and business context in one pass. Useful when you need to move from raw experiment data to an actual product decision without second-guessing the stats.