Photo via flickr
Usability testing is a cornerstone of effective user design. While usability testing is often a misunderstood part of the web and mobile design process, we wanted to highlight the importance of it. Our goal is to help everyone understand how UX designers create experiences that we all love.
What is usability testing?
Usability testing, or user testing, is when UX designers essentially sit down and talk to real life users about a product, service, or experience. By getting the user to actually weigh in on design decisions, it takes the guesswork out of design.
However, we humans are fickle things, and what we say is often completely disconnected from what we do. So to get around this problem, UX designers will often get a user to actually perform a series of tasks, talking out loud about their struggles and successes with a particular design.
For example, let’s say that a clothing retailer is getting some usability testing completed on the desktop version of its e-commerce site. A UX designer might come up with a series of tasks, each testing a different aspect of site design:
- Finding a particular item (testing Information Architecture (IA) and/or search)
- Finding out a particular piece of information about an item
- Contacting someone for help
- Following through on a purchase (testing purchase flow)
Each of these tasks covers dozens of micro-interactions, which together either make or break an experience. A UX designer will record where the user struggles, what goes really well, and from there can improve on the design.
The forms UX can take
What does UX actually look like? There are lots of variations out there, so we’re going to cover four of the most popular types, their pros and cons, and when to use them.
Remote Usability Testing
Remote usability testing is when the designer and the person are in different places, and the usability test is completed with some variation of web conferencing software. This can be a specific piece for usability testing, or it can be as simple as a Skype call. Other than that, the process remains much the same as our example above.
- The user gets to work in the real setting they’d actually use the program (e.g. their house, their office, out in the world if it’s an app).
- It’s often cheaper than in-person, because the designer can usually complete it at a desk or at home. All they need is a good internet connection, and maybe a decent set of headphones.
- Mo’ technology, mo’ problems. Web conferencing software can fail to install, users’ computers might reject it with a firewall, communication can cut out – the list of potential problems goes on.
- Sometimes designers can glean more insight from being there in person. Small body language doesn’t show up nearly as well over the internet, and designers might not be able to get the most out of a session.
It’s the same process as remote, just in-person. In person testing can be done anywhere, but it’s usually done in a formal research setting – one way mirrors, silent observers, a formal room and recording devices.
Photo via flickr
- Getting in front of the person can lead to better insights into whatever it is that’s being tested.
- In-person usability testing often provides the opportunity for stakeholders to see how their product is being used, which can be tremendously powerful and useful in making future decisions.
- It’s more expensive, which might mean less rounds of testing (thus less iteration on design).
Rapid Prototype Testing
This is a UX designers best friend. It can be done either online or in-person, but the idea is that early, early, early on in the design process, you get something in front of users.
The prototype can be as simple as a few screens, or even just sketches on various pieces of paper, with a designer manually changing the screen when the user ‘clicks’ on a page. (This is called paper prototype testing.)
- Can be done very quickly and very cheaply.
- Can be completed early on, when changes are still very easy to implement (pre-coding).
- Allows for lots of iteration throughout the design process.
- Advanced prototypes can take a long time to make, and be completely useless for the final product (although this problem is fading with new prototype software).
Focus groups are a little different. For one, there are lots of people – 6-9 users, according to Jakob Nielsen. It’s when a UX designer (or market research specialist – the line blurs) sits down with a group and talks about a project. Generally, they’re great for getting people talking about what they want a system to do, or how they want a product to work.
- Crystalizes what a system is supposed to do.
- Gets a lot of input from users quickly and effectively (since you’re getting nine opinions per session instead of just one).
- Good at outlining the problems with an existing system that users feel need to be solved.
- Particularly useful in the early strategy phases of a project, so that designers know what they’re aiming to do.
- Not very good at telling designers what the actual sticking points are. For example, everyone might agree in a focus group that they want a system to run faster, when in-person usability testing might reveal that it’s really number of steps in a process causing the frustration.
- If used without supporting testing, can lead to inaccurate data, because as we mentioned earlier, people tend to say one thing and actually do another.
Any user testing that you can do is beneficial. Any input from users, especially early on in a design process will help designers create better experiences. Remote, in-person, and rapid prototype testing are all great for looking at actual user behaviour. Focus groups are good for knowing what users want from a product, preferably before you start building.
But the best kind of user research? A combination, of course! By combining focus groups with rapid prototype testing early on, followed by in-person testing just before launch, companies can create an experience that their users want and enjoy.
If that’s not good UX, I don’t know what is.