Thinking about performing usability tests on your website is probably extremely overwhelming. I mean, I was slightly overwhelmed when Chris simply asked me to help him conduct simple usability tests on our own site, and all I had to do was read a script…literally. But something about the words “usability testing” incites this internal panic: How do we find the time? Where do we find the participants? Don’t we need a lot of fancy and technical equipment to do them? How much is this going to cost? If you’re not careful, you can come to the conclusion that it’s just not possible for your company to go down this road before you even explore the possibility.
Steve Krug’s book, Rocket Surgery Made Easy, is a great resource that completely dismantles the idea that usability testing must be a costly and time-consuming undertaking in order to yield advantageous results. In fact, if you take the right approach, usability testing can actually be really easy and fun. Imagine that.
Here are a few quick tips to help you get started with this process.
Tip 1: Finding the Right Person
For the type of usability tests that we’re recommending, it’s not always necessary to recruit users from your target audience. I’m sure this seems completely counterintuitive. After all, doesn’t it make sense that you would want to find volunteers that are representative of the people who will actually be using your site? Surely that would provide much more valuable insight! However, the types of problems that you’ll most likely uncover with these “quick and dirty” usability tests are ones that almost anyone would encounter—issues related to things like navigation, information architecture, and page layout.
In these cases, the ideal participant will be someone who is unfamiliar with the actual site you’re testing, but familiar with the web. Often, people within your company are too close to provide objective feedback—they typically know how the site is supposed to work already and have a personal investment in its success—but your 90 year old grandmother probably won’t be a fount of usability problem knowledge either. Consider recruiting friends or family. For our own tests, we recruited from our neighbors upstairs at BlogAds and even proposed a trade—they helped us with some usability tests and we served as their volunteers as well. And providing small incentives like a $20 gift card can sometimes sweeten the deal or at least show your heartfelt appreciation.
Tip 2: Setting the Right Scene
Make sure that you choose a quiet space for the testing room—this will minimize distractions and ensure that your recordings are free of background noise. In some of our recent tests, the participant was sitting in our main office—a wide open room where about 8 of us live Monday-Friday (during office hours only of course). While reviewing the tests, we noticed that we could hear the background conversations of other people in the room almost as clearly as the voice of the participants as they narrated us through their actions. My guess is that if it was distracting to us while watching the recordings, it was probably distracting to our volunteers as they made the recordings.
We recommend using just one screen for the tests. In the first round of testing we completed on our website, we recorded the participants using the webcam on a laptop, but the participants were actually using an external monitor to view our site and complete the tasks. This deprived us of a good, straight-on view of their faces. In his book, Krug actually recommends against recording the participant’s face, stating that it can be an unnecessary distraction. However, we found this to be helpful in our testing, as the facial expression of the participant provided some insight that would have been lost had we just been listening to an audio recording. If you find this added data helpful, I say go for it.
Before you start the tests, make sure you turn off or disable anything on the test computer that may interrupt the test—this includes instant messaging, email or Twitter alerts, calendar reminders, scheduled virus scans, etc. This will save you some potential embarrassment and allow the participant to remain solely focused on the tasks at hand.
We were able to learn from our mistakes and correct the noise and screen issues in a more recent round of tests. I’ve embedded a video from this month’s newsletter as a working example.
Tip 3: Always Be Ready To Do Another One
This one is pretty self-explanatory. Often you’ll encounter a problem during testing that will create more questions or a test will provide a surprise that will make you want to schedule another session to dig deeper. Go for it. As a rule, it’s a good idea to use three participants on each round of testing. But if you find that you’ll get valuable data by tweaking a certain task and scheduling another session, then by all means, go forth! Or if you feel that you just didn’t get enough data in the first three sessions and you want to schedule a few more, make it happen. By making the testing process nice and simple (and using tips 1 and 2), you should find that it’s relatively easy to add another participant or two to any round of testing.