Don’t trust your gut

We Ducks just wrapped up several website projects that involved user testing. It’s some of my favorite work, because the findings never fail to challenge me–or to remind me that my gut can only take me so far when it comes to building a good online experience.

So, how does user testing work? The basic idea is that we sit down with real people (the kind of the people who would actually visit the site), and watch them click around a website in response to prompts and tasks. Of course, a lot of rigor goes into trying to keep the experience consistent and avoid leading people with our questions, but at the core, it’s pretty simple. And it gives you a glimpse of how your visitors actually behave online.

I’m a keyboard shortcut girl myself, so when I watch someone painstakingly navigate to the “File” menu and choose “Print” from the dropdown, I’m reminded that the pace at which I experience the web isn’t universal. And after you watch the third or fourth person’s eyes pass right over that bright-colored banner you created to promote your big event, you’ll be ready to stop trusting your gut and start making some changes.

So if you really want to take your website to the next level, watch someone trying to use it. It doesn’t have to be complicated:

• Start by finding 4 or 5 people who reflect the diversity of your audience in terms of age, web savvy, socioeconomic status, interest in your cause, familiarity with your work, etc.
• Ask each person to sit down with you and go through the process of finding volunteer opportunities, or making a donation, or just figuring out what the heck it is you do.
• Watch their mouses (where do they click? what content did they miss?) and their faces (what confused them? what intrigued them?).
• Take notes on what you observe–if you can, you might even try recording it (we’ve used a service called Camtasia that works well) so that you can play it back later.
• Share your findings with a few colleagues, and spend 30 minutes brainstorming what changes you might make as a result of what you’ve found.

Even if you don’t do a formal process with experts, I can just about guarantee you’ll see something that you didn’t expect, or run up against an interesting challenge or a new idea for a solution. Try it, and tell me what you learn–or if you’ve done a process like this before, tell us what you found in the comments.

And in the meantime, if you want to learn more about your own biases and opinions about what works online, I suggest subscribing to WhichTestWon’s Test of the Week: every week, they’ll send you two different versions of a web page or email, and ask you to guess which one performed better. It’s surprising how often the best practice can be wrong depending on the context and the audience.