Today's keynote for DrupalCon 2011 couldn't have been more appropriate. Usability guru Jared Spool spoke on The Unintuitive Nature of Creating Intuitive Designs. Most web CMSs and related technologies struggle with this issue. If you want to do better, read on.

Manage the Knowledge Gap

An important concept when striving for intuitive design is the Knowledge Gap. Spool said to imagine an escalator going up. At the very top are the people who actually built the tool in question. Beneath them is the point called the Target Knowledge, or what the user needs to know to get their tasks done in this tool. Below that are the users, who have what's referred to as their Current Knowledge.

The space between Target Knowledge and Current Knowledge is known as the Knowledge Gap. There are two ways to deal with this gap: Training, or simplifying the design so that the Target Knowledge point is much closer to their Current Knowledge. For those who've been around long enough, Spool reminded us of all those little cardboard cheat sheets that used to come with programs such as WordPerfect, because there were so many keystroke commands that most people couldn't remember them. Then came the WYSIWYG editors that now dominate the market today.

Principles of Intuitive Design

In a talk full of examples and humor, he focused on a number of principles that help create intuitive design. The first is that an intuitive design lets users focus on a task, not the design. In other words, intuitive design is invisible.

As an example, he offered a U.S. Department of Agriculture site related to hay. There were just two options on the page, you could click "Got Hay" or "Need Hay." Makes total sense, right? But during the usability testing phase, they discovered that 50% of the users made the wrong selections. It turned out that people weren't sure whether "Got Hay" meant "I have hay to offer" or "Show me a list of who has hay."

The second principle revolved around understanding that "intuitive" is relative to the user in question. What does that person know? Something that's completely intuitive to a neurosurgeon, to how they've trained and how they work, would make absolutely no sense to an actor playing a neurosurgeon on TV. So, intuitive design is personal.

In an example that got a lot of laughs, he pointed out the fact that Drupal (news, site) founder Dries Buytaert said in his keynote that he'd create the Drupal 8 branch the moment he found out how to actually do that on GitHub. His point was that making something intuitive doesn't apply to just novice users.

Another important principle is that intuitive design focuses on user experience. Software often starts out simply, with some core features, and the Knowledge Gap is small. With each release, people add more features, making the Knowledge Gap wider. The problem is that often there's only a handful of features people really care about. Spool suggests you focus on those. If you do, then you're shifting from features to experience. In an ominous statement, he said that either you can shift from features to experience, or your competitors will.

The final principle he discussed was designing for embraceable change. Both American Airlines and JetBlue have ground to a halt in the past because they upgraded their systems to a "more useable design" in which the staff was completely lost. Spool asked us to imagine that, in the middle of the night, the usability fairies came into our homes and completely rearranged our kitchen, moving some things out of the room entirely, because they had "studies" that said there was a better way to place things. Our morning routines would be destroyed.

He points out that a user's goal isn't a better system. A user's goal is to accomplish things. Examples he offered of companies who handle change decently are Amazon, LinkedIn and Google, where each company has added bubbles or other features that announce that something has changed and explain what that means.

Measuring the Knowledge Gap

To understand how large your Knowledge Gap is, you have to figure out ways to measure it. The problem is not only that each user has their own knowledge, but there are many domains of knowledge and each user has different levels of knowledge within each domain.

So how do you measure this stuff? There are a number of techniques, some of which are quick:

  • Field visits: Grab members of your team and go play Jane Goodall, watching "users in the mist." Observe your users at their actual work and take copious notes. See how they actually work. Then adjust your design appropriately.
  • Usability testing: Watch people use your design and see where they struggle. Don't interfere. Ask questions when they're done.
  • Paper prototyping: Use paper to build mockups of workflow and design. Have your users use their fingers to show you where they'd click if they were using a mouse. Observe, record, report, adjust.
  • Five-second tests: Show a user your design for only five seconds. Then ask them to write down what they remember about it. Does what they remember match the purpose of that screen? If not, adjust.

Spool recommends watching your users for a minimum of two hours every six weeks, to help you constantly refine your design. If you're interested in learning more about this subject, he recommended two books: Handbook of Usability Testing, Second Edition, by Jeffrey Rubin and Dana Chisnell and Paper Prototyping by Carolyn Snyder.