Sex in a Box
“Wow! This software is sex in a box.”
The software in question was a hypertext application in the pre-Web era. When this participant clicked a link in the prototype and new content appeared, they had never seen anything like it before and were blown away.
Drinking While Driving
“This product is too complicated to use while drunk.”
The product in question was a device for NASCAR-race spectators. The feedback was from a field evaluation during which participants kept a diary of their thoughts while using the device during a race. Apparently, some participants were doing more than just watching the race.
Reconsidering Participatory Design
Some UX researchers believe that users are waiting for them with open arms, wanting to describe a day in their life, write on Post-It Notes, or test drive a Figma prototype. This is not always the case, as these stories from the 1990s show.
Too Many Changes
I was given access to some back-office staff so I could interview them for a project to enhance an internal system. There’s one interview I still remember decades later.
A fireplug of a woman sat down for her interview and said: “My boss told me to come to this meeting, but he didn’t say what it was about.”
I earnestly explained that we wanted her input on some enhancements to the software she used.
In response, she said: “My input is that I don’t want you to change anything. Every time you people change our system, we can no longer do our work the way we’re used to. So that’s my input: don’t change the system.”
Then she stood up and left the room.
I was working on a project for the Air Force of a certain European country to modernize a staff-management reporting system. When I sought access to the intended users, I had repeatedly been stymied. Eventually, someone took me aside to explain the reason.
The existing process was mind-bogglingly inefficient for a good reason: it provided jobs for lots of people doing their mandatory national service. These people had lucked into this office role, avoiding the typical experience of marching in the mud and having NCOs (Non-Commissioned Officers) yelling at them. Their primary goal was to complete their two-year commitment without jeopardizing their cushy office assignment.
Whether to Discard Data
Regrettably, user researchers sometimes face an agonizing decision about whether to exclude the data from a particular research participant. The following stories illustrate two schools of thought.
One guy started a usability session by picking up the mouse, holding it about 12 inches from his face, and squinting at it. It turned out that he had just come from an eye exam, and his eyes were fully dilated from eye drops. So I gave him his participant fee and told him that was all we needed for today.
I once had a participant who supposedly passed the screener that screened for people with a high level of computer literacy. He turned out to be a truck driver who had never used a desktop computer! He didn’t know how to use a mouse! What to do? Dismiss the participant or try to salvage the session? I’m a fan of salvaging whenever possible, so I ran the session. I’m glad I did because he gave us valuable feedback.
Surprise endings are not just for thrillers. Sometimes user-research sessions have them, too.
Slick or Usable?
A company once hired me to help choose the best software tool to buy for their employees. Working with some employees, we evaluated each candidate application and found that Product A was the most useful and usable—and Product C was the least. When I asked the employee evaluators which product they recommended, I was flabbergasted I heard their recommendation. They recommended Product C—the least usable and useful—because, they explained, it had the slickest-looking user interface.
Who’s the Expert?
Around the year 2000, I conducted usability testing to evaluate a tool that gave advice to IT administrators. The tool asked questions about their system configuration and offered advice on how to obtain better performance. Throughout the three-hour session, the participant was uniformly positive in his feedback. This was a little surprising because, around 2000, people did not generally believe that they could rely on computers to provide sound advice.
While chatting with the participant after the session, I said, “You seemed very positive, so I guess you’re looking forward to using this tool in your job.”
To my surprise, he replied: “Nah, I’m an expert who can do this task better than a computer can. Also, as an IT admin, I know computers are just dumb calculating machines, and I wouldn’t take advice from a calculator.”
This chance discussion cast all of the prior positive feedback in a very different light.
Fun fact—This tool was actually pretty smart. The people who developed it had earned PhDs and used it to set IT performance benchmark world records.
If you drew some profound lessons from these user stories, that’s great. If you had a good laugh, even better. I know there are many more great stories out there, so please share yours in the comments.
Acknowledgment—Several people contributed the stories in this article. Thanks to Bill Killam, Ilona Posner, Carl Zetie, Matt Belge, and others who wish to remain out of the limelight.