Description
This course equips managers to run ethical, credible A/B tests and interpret results without advanced statistics. You will learn to define success metrics, minimum detectable effects, and guardrails that protect the business. A planning worksheet walks you through sample size intuition and exposure limits in plain language. You will practice setting up hypotheses, variants, and holdouts and avoiding common pitfalls like peeking. The course includes a decision tree for when to test, when to ship, and when to run quasi-experiments. You will learn to translate results into rollout plans, risk assessments, and follow-on tests. A section on governance establishes who approves tests and how to share learnings across teams. Real examples show how conflicting metrics are handled and how to avoid over-optimizing a local maximum. You will receive templates for experiment briefs, tracking sheets, and readouts stakeholders trust. By the end, you will manage experiments that are fast, safe, and aligned to strategy.
Course + worksheets + experiment brief/readout templates
3 hours
A/B test planning, metrics selection, decision trees, governance, rollout communication.
Product managers, marketing leads, service owners introducing controlled tests.
