WHEN A/B TESTING DOESN'T WORK

Summary from the TLDR Web Dev newsletter

A/B tests are used to gauge the impact of simultaneous changes on different segments of a user base. While A/B tests can be highly beneficial for tech giants like Google or Meta, they often pose more challenges than advantages for other companies. Reasons for this include insufficient data, the tendency for incremental changes, the added workload of building and maintaining both features, and difficulty in determining what exactly to measure. Startups, in particular, benefit from being opinionated due to resource constraints and the inability to conduct extensive tests.