News for hackers. Stuff that matters.
Re:Ask HN: A/B tests on low traffic site – is it possible? by jonny383 on Thursday December 31, 2020 @05:38AM (#25591487)
If you flip a coin four times and get tails every time, do you conclude that flipping a coin isn't random?

Obvious answer: the more traffic, the better your results.

Re:Ask HN: A/B tests on low traffic site – is it possible? by faustas on Thursday December 31, 2020 @05:45AM (#25591526)
Yes, I know that. But there are many "suggestions" how to do AB with low traffic. Test significant changes, measure micro conversions instead of macro and etc. So I'm just wondering if all that is just "theoretical" things ir maybe there are some hacks that work in a real life...