Monday, January 15, 2018

Inexpensive Usability Testing


(Excerpted from my forthcoming book “Writing Effective Online Content Project Specifications”)

We rarely think of doing usability testing for online help or documentation (online content) because, after all, it’s just text. In fact, online content is text, but software forms a big part of it in the navigation, retrievability, and other features. And like any software, it should be reviewed for usability.

What if no group in your company does formal usability testing? Can you do it yourself without spending a lot of money? Yes. The results won’t be statistically significant but they will be useful and inexpensive. How to do this? In brief…

·         Create a prototype of your online content project. It can be a small but fully functional online output or a paper prototype. (If you’re not familiar with paper prototyping, see Paper Prototyping: The Fast and Easy Way to Design and Refine User Interfaces by Carolyn Snyder, published in 2003 and available on Amazon. The book is an excellent introduction to the subject, with additional points on usability in general.)

·         Create several little scenarios with questions whose answers can be found in the prototype. For example, “How many types of fixed asset depreciation are there?” or “What is the specific gravity of tungsten?”

·         Find a small but representative group of test subjects. If you’re testing a product for a general audience, you might set up your test in a conference room and invite people in at random to be test subjects. If your company runs a user conference, that’s also a good place to find subjects.

·         Tell the subjects that you need just fifteen minutes of their time. Offer a small gift (sometimes called swag or “tchotchkes”) as an inducement for them to help. (No pens or coffee cups – both badly overused.) For more impact, tell the subjects you’ll enter them in a raffle for dinner for two at a local restaurant.

·         Give the subjects the scenarios and ask them to find the answers. Watch them but don’t say anything or offer any help. You can tell them to follow the so-called “speak aloud protocol” where they’re free to say what they’re thinking. This may be a bit uncomfortable to hear but it’s the feedback that you want.

      The object is to see if the subjects can navigate through the online content to find the answers to the questions posed in the scenarios and how difficult and frustrating the process was. The results won’t be statistically significant, but may well be very enlightening. That’s what you want.