Is A B Testing Faster than Usability Testing at Getting Results

If HIV testing is faster and getting results than a usability test is, then should user research teams change their strategy, maybe even do less user research during our September virtual conference? Dr. Jacob Nielsen shares his thoughts on this question.

It’s an interesting tension here between the sort of almost immediate feedback you get from a B. Testing will only immediate if you have a large user pool, obviously, but if a lot of traffic, then you can pretty quickly get an AP test result and some people will say, well, that’s faster. We can just put it up. And two days later, we know as opposed to it may take two weeks to do the user research. But to me, I don’t actually really believe in this argument for first of all, it doesn’t have to take two weeks to do user research. Now, it probably does. If you are planning a study, doing a study, analyzing a study, producing the report. However, you can also have continuous use of research, always ongoing. This is actually one of the things I really like to recommend that very few people do. But if it’s Wednesday, it’s user day, and it’s like every Wednesday you have four people or five people come in or you have to set up appointments to go and have an online session with them, whatever. You don’t know in advance what you’re going to be testing. But one or two days before you poll, you know, your team. And what are your questions right now? What would you like to get some user data on? And that’s what you test that Wednesday. You know, it could be Thursday, whatever day it is, but just continuous, ongoing, preplanned. In that way, you’re not going to have to wait. You know, you can get your user data right away when you have a question. So that’s one thing it can actually use. It uses usability.

Studies can actually be done at almost any speed if you plan for to doesn’t have to be B slow. But the second one is, I think is a more serious problem, which is that a B testing is really in many ways a rather weak methodology for gaining insights. I mean, you and sometimes you get an answer, but you don’t get the much of the why. So you have to design variations. Right. And one goes does a little bit higher than the other. And so, yeah, the one that does higher is is the better one, but that’s only within those exact to design variations. And so if you don’t know why, then you haven’t learned some lessons that can apply to like the next when you were designing the next feature you’re designing of the next product you’re designing.

Leave a comment

Your email address will not be published. Required fields are marked *