When budgets are tight and clients are trying to keep costs down, “nice-to-have” services like usability testing are often the first to go when costing out a project. Over the past few months we’ve been trying out cheaper ways of running usability tests using online tools.
Why do usability testing at all?
We often introduce usability testing to our clients as an insurance policy. Yes, usability testing is an added cost that can seem daunting, but how much more expensive is it to develop and implement a website only to find that users can’t find that all-important “buy now” or “contact us” button?
In designing a user interface there are always assumptions made about user expectations and behavior. Hopefully these assumptions are well-grounded in research into the particular site’s objectives, the site’s audience, their goals and key tasks, as well as a sound knowledge of good usability practices. But all the research and best practices in the world aren’t the same as putting your site in front of real, live users and seeing what happens.
Facilitated usability testing = great results, but expensive
Traditional, facilitated usability testing is labour-intensive. Users are placed one-by-one in front of a prototype or an existing site while a trained facilitator observes and analyzes. The sessions are generally recorded for further in-depth analysis. This type of testing gives rich, in-depth results. Not only does it highlight potential problem areas on the site, it also provides insight into users’ thought processes and motivations. The facilitator can ask users about their choices, thoughts and frustrations on the spot. But what do you do if this approach just isn’t in your budget?
Remote usability testing to the rescue!
Now simpler, cheaper usability testing is possible using online testing tools such as Chalkmark and Usabilla. These tools simplify the testing process considerably. All you need to do is upload static jpeg mockups of your site design and a series of tasks for users to complete. Users log on to the website and are presented with the mockups and a series of questions:
The advantage of this type of testing is that instead of bringing in users one at a time and observing them one-on-one, all testing happens remotely. Users can take the test in the comfort of their own home or office. Thus, many more users can be tested in a fraction of the time, with no facilitator required. Sample sizes are considerably larger and you can test across a wide geographical region.
Results are compiled in the form of heatmaps which show where users clicked in response to each question. They provide easy at-a-glance results, letting you know which tasks users completed easily and which could be areas for concern. For instance, if users consistently click in the right place then all is well. If users are clicking everywhere, or in the wrong place then there is cause for concern.
So what’s the catch?
Like anything else in life, you do get what you pay for. Unfacilitated usability testing has great strengths but it also has its limitations.
What it’s good at:
- Verifying labels and terminology: Are the labels you used clear? Has the content been organized into categories that make sense to the user?
- Navigation, and calls to action: Can users make good guesses about where content can be found ? Can users complete key tasks on the site?
- Verifying design choices: Does the design help orient and guide users on the page?
What it’s not good at:
- Interaction: Because the test presents static jpegs only, interactive elements like rollovers, dropdowns, checkboxes, etc. cannot be thoroughly tested.
- Multi-step processes: Complex interactions such as search, check-out processes and forms are difficult to test.
- Finding out “why:”Because there is no facilitator sitting with the user, we don’t have the ability to probe the “why” of what users do.
Unfacilitated usability testing is a wonderfully useful tool as long as you are careful to respect its limitations. It’s a great way to check for gaps and verify assumptions before proceeding with development. And certainly any usability testing is better than no usability testing at all.
And, on a final note, as in any type of research, your test the results are only as good as the questions you ask; writing a good usability test plan is an art in and of itself. Stay tuned for more on this topic!
non-linear creations has developed test strategies and administered usability tests – facilitated and remote – for a wide variety of clients. For more information on this service or our approach to testing, don’t hesitate to get in touch.
