Ronny Kohavi's Selected Quora answers (upvotes 11/15/2019 used to sort)
- [112 upvotes] What is it like to be a data scientist at Microsoft?
- [76 upvotes] What are some good books to read on the practical side of online experimental design?
- [70 Upvotes] How long should you run an A/B test on your site before you declare one a winner?
- [40 upvotes] What common mistakes do companies make when running A/B tests?
- [34 upvotes] When should A/B testing not be trusted to make decisions?
- [33 upvotes] In A/B testing, what should be done if the results of an A/A test are statistically significant?
- [33 upvotes] What is your percentage of successful a/b tests or experiments out of all performed so far as a product manager?
- [30 upvotes] What are the levels of data scientists at Microsoft? How are data scientists at Microsoft promoted?
- [30 upvotes] How do unequal sample sizes affect A/B tests and analysis of the results?
- [25 upvotes] How would you run an A/B test if the observations are extremely right-skewed?
- [24 upvotes] Why A/B testing instead of A/B/C testing?
- [23 upvotes] How important is A/B testing to a data scientist working in a technology company?
- [23 upvotes] How do you verify that there is no bias in your sample when doing an A/B test?
- [20 upvotes] Can You Run A/b Tests On Things Other Than Web Pages?
- [20 upvotes] How small should the variations in my website design A/B test be?
- [19 upvotes] When I should not run an AB test?
- [18 upvotes] How would you design an experiment to determine the impact of latency on user engagement?
- [17 upvotes] How do you build and maintain an A/B testing roadmap?
- [14 upvotes] What are the pitfalls of A/B testing and what kind of alternative statistical tests can remedy these downsides?
- [13 upvotes] A/B testing: When are users not the right randomization unit for an experiment? What are the alternatives?
- [13 upvotes] What's the best way to AB test response times?
- [11 upvotes] Why isn’t population usually segmented in Internet A/B testing?