Designing for Behavior Change, 2nd Edition

Glossary of Terms

This book introduces many new concepts and terms, such as the behavioral map and the CREATE Action Funnel. This glossary provides definitions of those terms, for your convenience. For the sake of brevity, though, many of the common behavioral economics terms referenced in the book (e.g., implementation intentions) are not included here.

A/A test

an “experiment” in which there is no difference between the arms of the test. An A/A test allows you to verify that the randomization process and analysis code is working correctly. If you see a difference in outcomes across the two groups, there’s something wrong!

A/B test

a method of comparing two versions of something against each other to determine which one performs better. Two variants are shown to users at random, and a statistical analysis is used to determine which variation performs better for a given outcome metric. An A/B/C… test is an extended version, with 3+ versions.

A/Null test

a method of testing that takes the baseline intervention (aka, version of the product, feature, etc.) and tests it against no intervention at all. It provides the clearest and simplest measure of that intervention’s impact. It is also valuable to ensure that the intervention doesn’t actually make the user worse off (which is something that is missed with many A/B tests).

ability (per the CREATE Action Funnel)

a stage in the CREATE Action Funnel when the user evaluates whether or not she can take the target action now. ...