On UI: Forget About the 'Experts'
Peter returns to the topic of user interfaces to explain why you shouldn't waste any time helping your "expert" users -- and what you should be doing instead.
In responses to my previous user interface rant
, some readers pointed out how important it is that your UI support "experts."
No it isn't.
To a certain extent, I sympathize with the desire to provide an "expert mode." I often make fun of Amazon.com because I have to go through what seems like a dozen pages to buy a single book (most recently Zot! Book 1). But the right answer isn't to create an "expert" interface.
I had an epiphany about this when I was the head of an IT department for a heavy-equipment manufacturer. We created an online system that the company's customers could use to order parts at any time of the day or night, from anywhere in the world, without having to call one of our sales agents. (By "we," I mean the people in the department who actually did the work. My job as manager was to stand around and look pretty.)
Not lot of customers actually used our application. When asked, one of them put the problem very succinctly: "Why should I learn your silly-ass system when I could just call Bev?"
Why was our opinion about the system (which we thought was pretty cool) and our user's opinion so different?
You and Your Users
Stop for a minute and consider how different your user is from you.
You're an expert on your application...but you built the thing. How many of your users could be classified as "experts"? Consider: By definition, "experts" are a small part of your user community. You are, at best, similar to only a small part of your audience.
Now, consider how many of those experts are interacting with your application frequently enough to benefit from making the UI easier. While you've spent many hours with your application, your users probably haven't -- and won't. Cutting a minute off each interaction probably won't save your users any real time.
And, finally, consider your users' motivation. While you care greatly about your application, most of your users don't. They have no desire to learn a second advanced mode. Often, they're happy enough to learn exactly one mode for your application and go home.
Identifying with the Wrong Crowd
Let me tell you about another epiphany I had (and keep on having). I often end up doing usability testing with a "typical user" and taking the results back to the developer. Inevitably, the developer looks at the list of issues raised by the user and says, "Well, you must have gotten the stupidest person in the world to test my application."
Contrast this passionate desire to not support the typical user with the equally passionate desire to support experts.
I think the reason is obvious enough: We identify with the experts. As technical specialists, we value our own expertise and assume that our users do also. We're tired of our application and want to get through it as fast as possible; we assume our users feel the same way. We're willing to invest time in our application and assume our users are equally willing.
No, they're not.
The Right Answer
If you're going to worry about helping people, don't spend your time helping the experts. Help the typical user.
There's even a well-known pattern for what you should be trying to do: the Facade pattern. The goal of the Facade pattern is to make some complex set of objects easier to use by supporting the typical task. The Facade pattern doesn't consider whether you're an expert or not; it only considers what most users typically want to do with your application and makes that easier.
If you're looking for an example of the Facade pattern in action, you can look at Amazon.com. The people at Amazon recognized that when most users wanted to buy something, they wanted to buy it the same way they bought the last thing. So Amazon didn't worry about whether it's dealing with experts or not; it just added a button clearly labeled: "Buy now with one-click."
Now, that's a UI no one would call "silly-ass."
Peter Vogel is a system architect and principal in PH&V Information Services. PH&V provides full-stack consulting from UX design through object modeling to database design. Peter tweets about his VSM columns with the hashtag #vogelarticles. His blog posts on user experience design can be found at http://blog.learningtree.com/tag/ui/.