how we improved signups by 30% by doing nothing – the historious blog.
A/B testing is more-or-less a hot topic when it comes to the business side of design, in my opinion. It’s one of the few ways to get a quantitative result based on a design alteration. Or so we all hope.
One of the biggest problems with A/B testing that I can see is that in order to get a partially conclusive result, you need a lot of data. Unfortunate for us, the Web moves quite a bit faster than that, so I can only imagine how many times A/B tests were prematurely used to make a decision.
Lesson to be learned: keep an eye on your data, do everything you can to make sure it’s as accurate as possible. Even if it means simply waiting to collect more data.