Design experiments, estimate required sample sizes, evaluate detectability, and analyze results with statistical rigor.
Define metrics, hypotheses, traffic splits, power, and validate experiment design before running tests.
Estimate how many users you need before running an A/B test. Supports binary & continuous metrics, power, MDE, and traffic splits.
Find the smallest effect your experiment can reliably detect given your sample size, metric type, power, and test design.
Analyze experiment results, compare variants against control, identify winners, and get business recommendations.