The risks of dismissing minor usability issues

As a human factors psychologist, I am often asked to evaluate the usability of a product—its ease of use and ease of learning—to predict where users are going to run into problems, or, for products already on the market, to analyze why users are already having difficulties.

At the conclusion of these evaluations, I provide a report that details the issues I’ve found, categorized by severity: ranging from catastrophic—those that will cause the user to utterly fail at the task at hand, or, for medical and safety products, may even put the user or someone else in danger—to minor issues, those that should be fixed …if the time and resources are available.

But I’ve begun to question whether minor usability issues should be dismissed quite so readily. It’s true that a single seemingly trivial issue isn’t going to harm anyone, and a couple of them aren’t going to lead the user into utter failure. But minor usability issues seldom occur alone; they are the products of careless design and a lack of attention to detail, and when they occur, there are usually many such issues. Cumulatively, they can negatively impact your product in substantial ways.

Recently, as I was starting an interface evaluation, my initial reaction was, “wow, this is … not good”. But as I continued exploring the system, I realized that none of the problems I was encountering were ones that I would categorize as particularly severe. Rather the impression of a bad interface was being driven by a series of 9-10 badly designed microinteractions that I was encountering repeatedly.

They were simple things, like having windows with X buttons that didn’t actually close the window, whereas with other windows there was no way to close the window except by clicking the X. But the click target on the X was small and non-responsive… so when I got no reaction from the system, I was never really sure whether it was being slow to respond, if I’d missed clicking the X, or if the X was simply not working.

Another set of badly designed microinteractions consisted of error messages that were inconsistently handled and presented in a rather non-standard way (one that was distracting and intrusive).

Other controls were also also being used in non-standard ways, including checkboxes that appeared to be clickable, but weren’t. They were indications of states rather than controls, but they gave the mistaken impression that I could interact with them. And the checkboxes that were clickable had small target areas that were very easy to miss. So, again, when I clicked on a checkbox and nothing happened, I couldn’t be sure whether I had missed the checkbox or if I was clicking on one that I couldn’t actually change.

There were several other issues that I won’t go into, but you can see that individually, none of these was terrible and each was easy to work around, once I realized what was going on. And each of these issues might only have cost me a second or two. But this was a system that the users were expected to spend 8 hours on every day, and these badly designed microinteractions existed in dozens of windows. So, the user would encounter them hundreds of times a day, adding hundreds of 1-2 second hesitations, delays, and moments of confusion.

Each and every usability problem, no matter how minor, detracts from your product. Each and every one can reduce users’ performance, delay their responses, and increase their cognitive workload. When those minor usability issues add up, in safety critical systems they can dangerously distract your user. For a surgeon in the midst of a procedure, those minor issues could mean the difference between a complete success and a less-than-optimal patient outcome. For an industrial hygienist monitoring gas levels, those minor issues could delay a reading by just a few seconds … but what if that reading indicates that immediate evacuation is needed? For someone controlling a vehicle, those minor issues might take the user’s eyes off of the path ahead for just a fraction of a second longer. But what if that fraction of a second were the critical one?

Even in non-safety systems, those minor problems amplify your user’s frustration and diminish his or her confidence in your product. Each of those minor usability issues may just be a single straw … but we all know the proverb about the straw that broke the camel’s back. Are you willing to chance just how many minor issues your user will tolerate before they consider switching to your competitor?

Carolynn Johnson

Twitter Digg Delicious Stumbleupon Technorati Facebook Email