Usability Testing Notes

After all these years of building sites, I still breathe a quiet sigh of relief at the end of usability testing. Indeed, after awhile one develops a sense of what will and won’t work for various sites, but it’s always nice to know find that one has planned and executed well.

One of our clients just launched a resource site for parents in Illinois—a project for which we’d built in both wireframe testing and post-launch testing. Yesterday, we sat down individually with nine participants and had them complete various tasks and comment on the site as whole. Some good feedback fell under “things people are always happy about,” like “the bright and crisp photographs of people” and “clear and consistent navigation.” My favorite takeaways from this particular round of testing:

1. Interactive doesn’t always equal added value. Early in our planning, the client asked about creating an interactive map for all of the counties in Illinois-flashy rollovers and animation and such. I’d remembered that for another projectone involving countries and continents-we found that some people were hard-pressed to locate places like, uhm, Europe. That in mind, I wondered whether most people could rightly recognize their county. It seemed that the expense of programming that map outweighed its actual value, and the client agreed. I knew when one of the testing participants located Lake Michigan and Chicago on the left side of the state that we’d made the right decision. Sometimes, a low-fi alpha list just makes more sense. 

2. Except, when it does equal added value. People’s appetite for video-and their tolerance for lower production-quality stuff-has increased so dramatically. I was astounded and impressed with the requests for additional multimedia on the site. 

3. If you think it may be a little confusing, it’s really confusing. Both the client and I thought that certain acronyms-even with explanatory text around them-wouldn’t make sense to people outside the organization. We were right. Duh. 

4. Paper prototype testing still has value. Four years ago, I had client take me to task for recommending that we test wireframes—”SO late 90s,” she said. Well, now it’s late 00s and I still find wireframing has a lot of value. We tested wireframes for this site early on and came to clarity (at a fairly low cost) on the sorts of titles and site structures that would make the most sense to target users. As a result, the list of changes we need to make post-launch are few and mostly simple—nothing that requires a fundamental restructuring of the site. Paper prototyping, it’s a good thing. 

P.S. We’ve used MORAE software for years, but for those of us craving mac-native testing software, Silverback is wonderful. Give it a try if you haven’t seen it in action already.