A/B Test Setup

คำอธิบาย

A complete experimentation skill that helps you design A/B tests, multivariate tests, and full growth experimentation programs that produce statistically valid, decision-grade results — not vibes. It walks the user from a vague "should we test this?" to a written hypothesis, calculated sample size, defined primary / secondary / guardrail metrics, a pre-launch checklist, and a results read-out that doesn't fall for the peeking problem. Use it any time someone says "A/B test," "split test," "experiment," "multivariate test," "should I test this," "statistical significance," "how long should I run this test," "ICE score," "experiment velocity," or "experimentation program." It also covers the ongoing program: ICE prioritisation, hypothesis backlog, win-rate targets, and a playbook template that turns winning tests into reusable patterns.