Statistical Significance Checker for Low-Conversion Experiments

Assesses statistical power, sample size adequacy, and significance thresholds for A/B tests running in rare-event or high-variance environments — flags underpowered tests, peeking risk, and multiple-comparison inflation, then delivers a verdict with required sample size, minimum detectable effect, and a decision recommendation. Use when validating whether a low-conversion or fintech/e-commerce experiment has enough data to declare a winner, or when a result looks significant but the base rate is tiny.