Demystifying the Dreaded A/B Test
summary
Starts with a concrete claim: target retention will increase by 20% and the goal is to hit that target not overshoot it. Uses Split Software feature flags, a vanilla JavaScript SDK or pure CSS toggles, and simple segmentation like 50% split or sending B to 10,000 people in San Francisco. Shows a live CSS refresh demo that alternates A/B, gathers metrics, then picks the winner from analytics and rolls the feature out.
AI-written summary · about
chapters (6)
- 0:00Intro
- 1:35What is A/B testing?
- 2:40Why?
- 4:34First, gather the data
- 6:04Feature flags
- 6:47Flex that UI muscle!
related talks
32mAB Testing to Determine Product Priority
45mHow we killed 80% of features and increased outcomes of a/b testing by 100%
38mTesting, Testing: On Experimental Drift and Data Driven Product Design
24mDoes A/B Testing Work?
27mA Tale of Two Components: Mastering A:B Testing in React
28mOptimizing your A/B testing roadmap
33mHow to A/B test with confidence
25mMake a Difference with Simple A/B Testing