So far, our user research has been pretty theoretical. We've painted a statistical portrait of our users and created fictional characters to personify them. We've talked to users about their wants and needs. That's all well and good, but what you really need to know is how (and whether) they're going to use your site. traffic analysisThe best way to understand user behavior (and perhaps the best way to understand users, period) is watching how they actually use your site. How long do they stay? Where do they click? Do they make a purchase? Find important features? Only your traffic logs know. You can learn a great deal from your traffic logs about who your users are and how they interact with your site. Traffic analysis can tell you how they arrived at your site (from a search engine, through a link on another site, etc.), how long they stayed, where they clicked, and the features or pages they chose. (See monitoring & evolving your site, p. 240.)
Traffic analysis and usability testing work hand in hand to show you how your site is used. Log files can also reveal "user clues" hints about their identities or backgrounds such as their host domains (i.e. intel.com, stanford.edu), the times they visit, and their computer systems and browsers. Traffic analysis and usability testing work hand in hand to illuminate your site's potential pitfalls (and its stunning successes). Traffic analysis presents facts, but can raise more questions than it answers. For example, your log files may reveal that 90% of your users leave your site after the first page. But logs don't tell you why. Is the site too slow? Is it malfunctioning? Is the design confusing? Did users end up there by mistake? What?! Usability tests attempt to answer these questions by shedding light on the user's thought process during a typical session.
usability testingUsability testing gives you a chance to see real people use your site in a way that simulates the actual user experience while allowing for questions. Do they understand what the labels mean, where the links lead, what the buttons do? Does the interface work the way they think it will work? Can they complete a given task? Nothing is as powerful or as instructive as watching real users struggle with your interface. And struggle they will. No matter how good your initial design or how talented your design team, some aspect of your interface the navigation system, the labels, the sequencing of events will likely confuse your users. (See designing for the user, p. 82.)
By putting your site in front of users early in the design process, you can catch significant problems early enough to correct them. For usability testing sometimes turns up problems that are more than skin deep. You may discover problems with your site's functionality or conceptual structure that force you to rethink your approach.
Listen to what they tell you. Customer feedback is the most direct path you have to users' brains. So the greatest favor you can do for your users and your product team is to integrate usability testing into every step of the design process. You should begin testing as soon as you have a prototype on paper just sketches of the interface, really and continue testing until you complete the site. Test early and often (at least twice before launch); this is the best way to catch mistakes and avert a user disaster. This might sound complicated and expensive, but it need not be. The beauty of usability testing is that it can fit any budget. Of course, it's nice to have a state-of-the art usability lab and a full-time testing and recruiting staff. But usability testing can be done competently with a camcorder, a part-time tester (you, perhaps?), and a small budget to compensate volunteers. In a pinch, you can even test the site with friends and family. As usability expert Steve Krug writes in Don't Make Me Think, "Testing one user is 100% better than testing none. Testing one user early in the process is better than testing 50 near the end." customer feedbackThe final clue to understanding user behavior is also the most obvious: Listen to what they tell you. Customer feedback email, phone calls, or actual letters is the most direct path you have to user's brains. It's honest, candid, and importantly, it's proactive: Brought to you by the users themselves, rather than offered in response to your questions. Now bear in mind that customer feedback always skews negative (people are more likely to let you know when they're outraged than when they're delighted), so it isn't an accurate gauge of overall consumer feeling. It is, however, an accurate portrayal of user frustration. You can safely assume that every angry email you receive represents the experience of at least 10 users nine of whom simply left in a huff without bothering to tell you that you'd lost their business. So it's important to pay attention to what the cranky users are telling you. People don't usually write in about run-of-the-mill frustrations slightly confusing design, slightly annoying features, somewhat slow pages they write when something has them fuming. So it's important to give credence to these letters, even when they're written in a rather irrational tone. A wise manager views customer feedback as a valuable research tool. And also an opportunity. Customer feedback gives you a chance to turn angry customers into loyal customers. These cranky letter-writers are very vocal, active people who tell people what's on their mind. If you win them back, you've gained a very powerful advocate. One more note on customer feedback: If you have a large enough organization, and you're high enough up the ladder, it's easy to be shielded from the experience and reactions of actual users. But it's important to expose yourself now and then. You need to understand: What confuses your users? What do they like? What do they hate? What frustrates them? And why?
|