Usability testing is an important part of the web development process, but it’s also one of the least-well understood. Very often we see misconceptions about what usability testing is, and what it isn’t.
This blog post will hopefully provide a primer on how you should be using usability testing.
One of the most important concepts in usability testing is that it should focus on how easily the user can complete a given task.
These tasks should be clearly defined from the outset of the project, and should be the key actions that you want your users to be able to perform. For example, this might be:
- Filling in a contact form
- Finding essential contact information
- Completing a purchase
It’s essential that we know from the outset what the goals for your website will be; this allows us to create appropriate calls to action and a design that is consistent with these goals.
When testing how well users complete these goals, there are four criteria used to judge the success of the test:
- Performance: How long did it take the user to complete the task? How many steps did it take?
- Accuracy: Did the user make mistakes in the process? If they did, were they able to recover from them?
- Recall: Could the user remember how to complete the task afterwards? Will they remember how to do it in the future?
- Emotional response: How did the user feel on completing the task? Stressed? Relieved? Comfortable?
Depending on your budget, there are various methods for implementing usability testing.
Hallway Testing
The least expensive option is hallway testing, where people unconnected with the project are asked to complete tasks, as specified above. The advantage of this method is its cost. This method is also called “hallway intercept testing” – literally stopping people in the hallway and asking for their help!
The disadvantage of this method is that it doesn’t necessarily provide a good cross-section of the group that will be using the site. It’s an informal method, and the familiarity of the user and the person conducting the tests could influence the outcome of the tests.
Remote Testing
Remote testing involves the use of specially set up online surveys to test the usability of a site or application. One such service, Amazon’s Mechanical Turk, offers a cost-effective method of testing your site with a wide variety of users.
Although more expensive than hallway testing, this method offers a more formal testing structure, and encourages a goal-centred approach. With Mechanical Turk, you can also choose to reject submitted results if you feel the quality of feedback is not good enough.
Expert Review
The third, and (by far!) most expensive option is expert review. This involves working with a specialist usability testing company to provide in-depth analysis for the goals listed above. 360innovate work with a number of these companies, and can suggest one that is suitable for your needs.
Usability testing: what it’s not
The most common problem we see with usability testing is that it is performed without goals in mind, or that the goals are not clearly defined at the start of the project. In effect, test subjects are being asked “what do you think of this site”?
This is very definitely not usability testing. Unless you relate feedback to the original goals of the project, it’s very unlikely that you’ll get useful feedback on the success of the project. You might get some useful information for future development, but you won’t get useful feedback on the success of what you’ve built.
In short, you should know what the goals of the project are before the first the first line of code is written, and you should be testing how well each task relating to a goal has been implemented.
Why should I bother at all?
Usability testing is a vital tool to judge the success of a project. The internet is awash with stories emphasising the importance of usability testing, and the real question is “can you afford not to perform usability testing?”
Please get in touch if you’d like to talk about what usability testing can do for you.