A/B testing for 250 million people
summary
Indeed, the #1 job search site with ~250 million monthly unique visitors, describes using A/B testing to validate every new feature and product change. The talk covers why A/B tests are needed, how Indeed’s A/B testing framework runs tests and handles feature splits, how to test the tests themselves, and how to analyze test results.
AI-written summary · about
chapters (11)
- 0:00A/B testing intro
- 3:27Basics and multivariate
- 7:19Benefits in development
- 10:08Prowler platform
- 14:51Test creation process
- 21:06Large project challenges
- 25:42Automation and scheduling
- 27:51Metrics and logging
- 31:44Best practices
- 33:08Resources and closing
- 34:13Q&A
related talks
32mAB Testing to Determine Product Priority
24mDoes A/B Testing Work?
28mOptimizing your A/B testing roadmap
33mHow to A/B test with confidence
28mBeyond A/B: Test automation approaches for web variants from A to Z
09mDemystifying the Dreaded A/B Test
45mHow we killed 80% of features and increased outcomes of a/b testing by 100%
43mLean QA: Automation, Monitoring, Experiments