Your new website is up and running. You’ve agonized over the font, the color, the photos. But have you actually sat down with some of your users and watched how they use your website? If not, you may be in for a few surprises.
User testing is one of the most important steps you can take to create a successful website. And, contrary to what you might think, it doesn’t require a huge budget, fancy designers, focus groups, or in-depth data analysis. In fact, you can put together an effective user test with just five people.
Website usability testing has been on our minds lately, in part because it played an important role in developing our new improved user interface. We realized that the same lessons we were applying to Jimdo at large could be adapted by everyone with their own websites. I spoke to Enzo Pietzsch, who leads Jimdo’s usability tests, to get some of his tips on conducting your own “DIY” version at home:
What is website usability testing?
In a basic sense, usability (or user) testing means sitting down with someone and watching them use your website while they describe out loud what they are thinking.
Most of the time, you ask them to perform a few tasks—sign up for a newsletter, choose an item in the online store— and you observe what, if any, issues they come across, like dead ends, confusing navigation, or unclear directions.
Enzo tells me that he tries to conduct basic usability testing about every two weeks, with testers either from UserTesting.com or from his own pool of volunteers. “A usability test is a great tool when it comes to understanding how users interact with your site,” he says. “It also helps you avoid user frustration because of something not being self-explanatory and intuitive.”
What usability testing can show you
Usability testing is not about gathering people’s opinion of your website, since those can’t be accurately represented by such a small sample of people. It’s about observing the actual steps people take to get from Point A to Point B. This is important because what people say they do and what they actually do are often very different. Watching a user go through a set of tasks can shed light on how they use your site in the real world.
Why is this so valuable? As the website creator, you’re so close to the material that you might not be able to spot potential problems. You know where the “Contact Us” page is, so why would other people have trouble finding it? But your users aren’t you, and they can get tripped up in unexpected ways.
Even established websites can learn a lot from user testing. Check out the following video from a Spotify user test and you’ll see what I mean.
Website usability testing the do-it-yourself way
For those with large websites, time, and budget, there is a formal approach to usability testing, often involving more extensive tests. For most people, though, this approach is overkill. “You don’t have to do it by the book to get useful data,” says Dana Chisnell, who actually did write the book on usability testing. There’s a basic methodology, but even informal approaches can be extremely useful.
One low-cost option is to use a service like UserTesting.com. You design a set of tasks and choose your target audience, and they provide the testers. Most tests cost a few hundred dollars, depending on the number of participants you want, and you can remotely look over people’s shoulders and hear their responses as they use your site. You can also request a free trial to see how it works.
If that’s outside your budget, another option is the DIY route. You recruit your own testers and simply sit next to them while they try your site and go through a few simple tasks. If your users are remote, you can set up a GoToMeeting or use Google Hangouts so they can share their screens with you while they talk through what they are doing.
Choosing the right users to test
The basic rule of thumb is that you can do simple tests with about five people, spending about 15 minutes with each person. Too much more than that, and you might get more feedback than you can handle. But even the five-person rule is not etched in stone. “A test with only one person is 100% better than no test at all,” says Enzo.
While it’s a good idea to test with your customers or target audience, Enzo recommends not stressing too much about finding the “right” testers. “I think ‘target audience’ is often overrated in usability tests,” he says. “Testers should meet the basic characteristics of your audience, but anything more specific, like peer group or income, isn’t that relevant since what you’re testing is universal behavior.” So go ahead and test on your mom, your friends, anyone who you feel comfortable with.
The exception would be if you are testing a feature for a particular operating system like Android or iOS. In this case, you would want to test with people who are already familiar with those platforms.
Designing the right usability tests and tasks
Each test you conduct should have a goal in mind; you don’t want testers just meandering around your site randomly clicking. Focus on a few top-priority parts of your site:
- Anything that a typical user would want to do right away when visiting your website, such as finding your opening hours or contact information. Problems here could make users leave your site or give them a negative first impression.
- Anything that is critical to the purpose of the site. For example, if e-commerce is a major component of your website, ask users to complete a purchase in your online store. If it’s really important that they can view your portfolio, design a task that asks them to find examples of your work.
- Anything that would help users “seal the deal” – i.e. pay for an item or sign up for a newsletter. If a person gets stuck at these key moments, it doesn’t matter how nice your site looks—the website won’t do its job.
Also, remember not to give away clues in your tasks. For example, rather than saying “Use the navigation bar to find the ‘Contact Us’ page,” say, “If you wanted to get in touch with us, what would you do?”
Some experts say that they get the best results by presenting the tasks as “scenarios” that give the character, the context, and the goal. For example, you could say, “You are planning a last-minute dinner party and you need to find a recipe that is vegetarian and suitable for kids,” rather than “Find a vegetarian recipe.” Creating a story, even a very simple one, helps users relate to the task at hand, and makes them more likely to react in a natural way, which is exactly what you want.
Test and test again
You don’t need a lot of people, a lot of time, or a lot of money to do some basic user testing. Usability experts will tell you, however, that many small tests are more effective than one big test. Testing early and often, throughout the website development process, will save you a lot of time and effort later on. “It’s important to start early,” says Enzo, “so that you don’t get attached to a suboptimal solution.”
The great thing about usability testing is that it can be as simple or involved as you want it to be. Either way, you will likely gather some surprising insights into how your website works in the real world, with real people. The number of participants, the questions you ask, and the number of tests you run are all up to you and can change depending on your time and budget. The important thing, though, is to actually get started and do it.
Anything surprising you learned from your own usability tests? Let us know in the comments.