Blog

Donnu A/B Blog

Complete, honest guides on CRO, A/B testing and experimentation. The most complete articles on the web, no fluff. (page 2 of 17)

Statistics

Equivalence Testing: Proving a Variant Did Not Hurt

Equivalence testing proves a variant did not hurt. How to set the margin, read the 90 percent interval and size an A/B test to establish a tie.

Statistics

Futility Stopping: When to Shut Down an A/B Test

Futility stopping uses conditional power to shut down an A/B test that cannot win. How to compute it, where to put the threshold and what the error costs.

Statistics

Observed Power: The Number That Answers Nothing

Observed power after an A/B test is the p-value rewritten. Why it tells you nothing about a non-significant result, and what to read instead.

Statistics

Bot Traffic in A/B Tests: Clean Before You Read

Bot traffic only breaks an A/B test when it lands unevenly across variants. How to tell noise from bias, detect it with A/A tests and filter it.

Statistics

Regression to the Mean: Why the Peak Never Repeats

Regression to the mean makes the page you picked for being worst improve on its own. How to separate statistical recovery from a real A/B test effect.

Statistics

Tracking Loss in A/B Tests: The Data That Vanishes

Consent, blockers and browser policy erase part of your data. Tracking loss only biases an A/B test when it hits one variant harder than the other.

Statistics

Heterogeneous Treatment Effects in A/B Testing

Heterogeneous treatment effects: a significant result on mobile and none on desktop does not prove the effect differs. The interaction test explained.

Statistics

Multiple Metrics in One A/B Test: Controlling FDR

Reading 20 metrics in one A/B test gives a 64% chance of at least one false positive. How Bonferroni and Benjamini-Hochberg control multiple metrics.

Statistics

Sample Size for Revenue and Continuous Metrics

Sample size for revenue per user comes from the coefficient of variation, not the conversion rate. The continuous formula and the common mistake.