Designing the Obvious, Part 6: Usability Testing
You’ve heard all the hype. You may have read a book about it. You may have even done it once or twice. But if you’re like most people, you haven’t dedicated the time and energy to it that you probably should, and your products are suffering for it.
I’m talking, of course, about usability testing.
If you haven’t done it, it’s probably because it sounds incredibly time-consuming, expensive, and complicated. Usability testing is usually explained in such a way that the "rules" for doing it are vague at best, and it’s not an interesting enough subject to devote a lot of time finessing the details to become a master, so you don’t bother with it at all.
I understand. I’ve ignored it myself on countless occasions. But I’ve picked up a few tricks to make usability testing fast, cheap, and easy, and they just might convince you that it’s worth your while.
When to Stop Using Personas and Start Using People
In a previous article in this series, I talked about using personas to capture the goals of your users so you could better determine what to build that would satisfy their needs. And while it can be a very useful tool when you have the time to perform thorough user research, some people advocate using these personas in a low-grade form of usability testing known as scenarios.
Scenarios are basically short stories about how your key personas would use and interact with your product. The object is to write about the real-world tasks your product is expected to handle, and see how the personas might deal with each task. This may sound a little strange—and honestly, it probably should, because it is. Personas are imaginary characters, and while they’re based on real users and are meant to represent the target audience for your product, there’s no getting around the fact that they’re imaginary. Similarly, scenarios are fictitious accounts of hypothetical situations in which your imaginary characters interact with an imaginary interface.
As you can see, the air can get pretty thin up there in HappyCloudLand. We’re web geeks, not character actors. Testing imaginary characters against imaginary interfaces is not our forte. My take on the situation is that once you have defined a feature set, based on personas or through some other method, the personas have outgrown their usefulness. When this happens, it’s time to move toward something more tangible: real users.
Usability testing can offer many benefits that scenarios can’t. Getting real people to use your designs and give you feedback on them is a fantastic method for surfacing problems with an interface.