Day 0	Ship to 25% of users. Watch: errors, support contacts, the guardrails.
Day 3	If nothing has broken, go to 50%. Check the primary metric is behaving as the test predicted. If it isn't, that's worth pausing for.
Day 7	Go to 100%. Remove the old code path once you're confident.
Week 4	Check the effect is still there. A gain that has faded was probably novelty.
Week 12	Check the lagging metric your primary was a proxy for. This is the step everyone skips and it's the one that tells you whether the proxy was ever worth trusting.

The Week-Twelve Check Is the Valuable One

If your primary metric was a proxy for something longer-term, this is where you find out whether the proxy held. Over several tests it also tells you whether your entire measurement approach is sound. Almost nobody does it, which is why teams can optimise a proxy for two years before discovering it was only loosely connected to the outcome.

A/B Testing for Product Managers

Module 1Why Most A/B Tests Waste Three Weeks

Module 2Is This Even Worth Testing?

Module 3Writing a Hypothesis You Can Be Wrong About

Module 4Choosing the Metric

Module 5How Long It Needs to Run

Module 6Running It Without Ruining It

Module 7Reading the Result

Module 8Deciding and Shipping

Module 9The Six Ways Tests Lie

Module 10When A/B Testing Is the Wrong Tool

Module 11Final Project: Design, Run and Report One Real Test

Module 12Appendix: The Prompt Library

Module 13Glossary: Words Explained

Rolling Out