Once you’ve done all of the things you can to make your website easy to use, effective, and efficient, it’s time to watch and learn. (Unfortunately, unlike Green Cove Springs, we do not offer pre-lesson shooters. Sorry.)
A great way to test your site’s usability is to simply sit behind someone who’s never been to your site before and watch him or her click around.
Notice where your user goes and how he or she gets there. Don’t point things out or help the person to navigate. Just observe! (Pay extra close attention to the first few seconds.)
You are likely to be surprised about…
…How people use your site (or misuse your site, as the case may be).
It’s natural to assume that your site is intuitive and easy to navigate. Keep in mind, though, that someone visiting for your site for the first time doesn’t know what you know. The things that you think are “obvious” may not be to your average visitor.
After your observation, ask yourself the following three questions:
- Was it easy for your visitor to get to what he or she wanted to get to?
- Were site elements located where your visitor expected them to be?
- Was the visitor slow to navigate the homepage (i.e. was he or she overwhelmed by the amount of content or options)?
Once you get deeper into this exercise, it may be easy to get distracted by issues that occur on other pages of the site. It’s important to note these, of course, but it’s essential to start on the homepage.
If your site is not scannable and immediately understood, a visitor will click elsewhere. So, first, nail down the key elements in a user’s first few seconds that need to be improved and, then, attend to the rest of your list.
If you are unsure of how to conduct a usability test, watch this 24-minute video from author and usability guru Steve Krug. He recorded this demo test as a companion piece to his latest book, Rocket Surgery Made Easy: The Do-It-Yourself Guide to Finding and Fixing Usability Problems. His main purpose for creating this video is to demonstrate how simple usability testing can and should be.
