What problem does it solve? Teams often run experiments without clear hypotheses, adequate sample sizes, or disciplined analysis, leading to false positives and wasted traffic. This Skill guides the design, execution, and analysis of A/B tests and builds a continuous experimentation program with statistically valid, actionable results. ## Core Features & Use Cases - Hypothesis-Driven Test Design: Structures hypotheses with observation, predicted outcome, audience, and measurable metrics, then selects the right test type (A/B, A/B/n, MVT, split URL). - Sample Size & Duration Planning: Provides quick-reference tables, duration formulas, and sequential testing guidance to prevent underpowered tests and the peeking problem. - Growth Experimentation Program: Covers ICE prioritization, experiment velocity tracking, cadence rituals, and a playbook for compounding winning patterns. - Use Case: A team wants to test a new pricing page headline. The Skill calculates required sample size from baseline conversion and traffic, defines primary/secondary/guardrail metrics, and produces a documented test plan. ## Quick Start Ask the assistant to design an A/B test for a specific page change, providing your current conversion rate and monthly traffic.
