Most teams don't have enough traffic to test most things, and this is rarely said out loud. The alternatives are legitimate methods, not consolation prizes.

Instead of a Test	How It Works	What You Give Up
Watch five users	Structured observation of people doing the task	Quantification, you learn why, not how much
Ship and watch	Release to everyone, monitor the metric for a fortnight	Can't separate your change from everything else
Staged rollout	Release gradually, watching for problems at each stage	A proper comparison, but you catch disasters
Before and after	Compare the period before to the period after	Seasonality and other changes confound it
Test a bolder change	Make a bigger change so the effect is detectable	Precision about what specifically worked
Test on the highest-traffic surface	Move the test to where the volume is	You're testing a different thing

Bolder Changes Are Underrated

If your traffic only supports detecting a 15% lift, then small refinements are untestable and you should stop proposing them. That constraint pushes you toward bigger, more distinctive changes, which are usually more interesting anyway, and which produce clearer learning whichever way they go.

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

What to Do With Too Little Traffic