A/B Testings
A/B testings are the big hype right now.
A top well known Silicon Valley investor a few days ago put A/B Testing firm Optimizely on the cusp of his startup watch.
Today I read that the Huffington Post is A/B testing their headlines and it is almost presented as the most disruptive journalist practice ever.
I’m happy that something related to user experience is getting so much attention but… when I read things like this…
From direct mail to web design, A/B testing is considered a gold standard of user research: Show one version to half your audience and another version to the other half; compare results, and adjust accordingly.
It looks so simple and all the credits go to the technique, nothing to the designers behind.
What if you have 2 versions of a homepage? A is just bad and B is a total piece of crap. You A/B test them and A doubles the statistics of B. You deliver a report to your boss. His proud of you because you are using such a cutting-edge techniques (he reads TechCrunch), and he is also happy because the homepage finally implemented is giving 100% better results than the other.
At the end of the day, everybody can happily go to sleep.
Except the users of that site of course…
A/B testing is a good technique. Usability testing, participatory design, card sorting, etc, are also great techniques when properly used. But none of them assures you a good product.
It takes a good designer to get a good design. Period.
