Chapter 1. Introduction: Nothing to Hide

4 Where We Got Our Data

• How We Did the Book Study

• The "Thinking Aloud" Method

• Site-Specific Testing

• Sites Tested

• Web-Wide Testing

• What if a Site Has Changed?

17 Tell Me Again: Why Do I Need to Do User Testing?

• User Testing in Three Days

• The Exceptions

We'll begin by letting you in on a secret: There are no secrets to doing good usability studies. It's simply a matter of knowing how and where to look, and then documenting your observations.

Usability works because it reveals how the world works. Once you discover how people interact with your design, you can make it better than your competitor's.

This book tells you how people behave on the Internet and what makes Web sites fail or succeed. How do we know this? Because we've spent hundreds of hours looking. If anything, that's our big secret: that we know how to conduct user research. Yet even that is not a secret because we teach a course on how to conduct usability studies correctly.

As much as we try to teach companies to do their own user research and to employ valid methods, however, we keep getting asked for advice on how to make Web sites better. Many people would rather know what works in general than spend time testing their own designs. With so much already known and documented about user behavior on the Web, why not base your design on this general knowledge? Then you can always fine-tune it by conducting research on industry-specific questions that apply to your particular site.

We are writing for sites that are not among the Web's Top Ten most-visited list. It's actually good news if you're not on the Top Ten most-visited list.

Fair enough. That's why we wrote this bookto bring the best of our extensive knowledge on general usability together in one handy place. The observations here are based on issues that we have seen again and again with many different users as they have tried to use many different Web sites. The guidelines translate our empirical observations into general advice about what usually works on the Web. Although there are exceptionswhich is why we advise that you test your own sitethese guidelines apply about 90 percent of the time, and the vast majority of Web sites would be better if they complied with them.