Why Early Usability Testing?

Tobias Komischke / Monday, January 13, 2014

Usability testing is observing how users interact with a technical system, a product or a service in order to learn how this system can be optimized for a better user experience. While usability tests should be run at various phases during a system’s lifecycle, they’re especially valuable in the early stages - before the design is complete and before we do any development. At this point there’s no prototype built in code on the actual target development platform, but that’s OK. Usability tests can be run with paper prototypes showing static screen states. I personally prefer something that is more interactive, like linked PowerPoint slides or Indigo Studio mockups. The goal of this kind of mockup is not to be fully functional and real, but to mimic, to some extent, what the envisioned product would be like. We put as much in the mockup as necessary to make the underlying design tangible for users and as little as possible in order to keep the time and costs low. What we want to discover through the testing are things like:

· Do end users understand what they’re looking at?

· Do they understand the labels we use?

· Do they understand how to find their way around in the product?

· Is it clear what to enter in specific text boxes?

· Does the UI support the tasks sufficiently?

· Is the level of user guidance appropriate?

The mockup is only built to support these types of issues. It’s not meant to be complete, either horizontally (range of features) or vertically (feature depth). It’s also not meant to investigate nuanced things like animations between screen states. It does not have to allow execution of complete work flows. Since the users are instructed to complete certain specific tasks only, it becomes the role of the moderator to steer test participants to those mockup areas we’re most interested in and away from areas that are incomplete.

Would it be great to test a 360 degree view of the whole product? Yes it would, but to accomplish that we’d pretty much have to build the real thing. User acceptance testing would cover that, but only very late in the development process. At this point, when we’re still in the early stages of this process, we need to focus on the most important things. And of course, it’s an option to run further usability tests later, that focus on other areas of interest and that allow more granular interactivity.

What we learn from testing is very valuable for the whole development process. For the first time we get feedback from the only group of persons that is actually qualified to rate the usability of the product under construction –the end users. The usability tests allow us to assess the validity of our concept: we understand what works well and we’re able to pinpoint areas where the usability should be improved. Thus, usability testing is not a waste of time and money. Quite the opposite: it helps save money because we identify issues early and fix them now rather than later, after development, when changes are much more costly. The table below provides some numbers substantiating this claim.


Image showing estimates of relative Cost of Rework

Remember: there are a lot of shapes and flavors of usability testing (formative vs. summative, remote vs. onsite, moderated vs. unmoderated, etc.), but the most important thing is to actually do it.