From “usability testing” to “user-centered design”

Tagged:  •    •  

Many libraries employ usability testing to assess new or updated online services, such as database listings, catalogs, and digital collections. A popular model is observing a small number of potential users perform prescribed tasks using the service, and noting where those users encounter problems. This model, when used as the sole method for gaining user data, requires that the service be relatively well-developed before testing is done, increasing the risk that the fundamental design of the service does not meet user needs.

Task-based testing is most effective when used as one of a suite of tools ensuring library services are designed around users. An individual trained in various methods for identifying user needs and analyzing how well a service meets them, and knowledgeable about user behavior in general, is an essential member of the design team for new or updated online library services. That individual might choose from any number of methods to gain user data to answer a given design question, including (but certainly not limited to!), heuristic evaluation, focus groups, surveys, anthropological methods such as those employed by the University of Rochester, log analysis, card sorts, paper prototype walkthroughs, and the aforementioned task scenarios. Significant expertise is needed to match a methodology to the question being asked, and to interpret the often qualitative and sometimes conflicting data into an evidence-based decision. An effective voice for user-centered design thinks outside the capabilities of existing software and asks the common-sense questions users will ask despite existing institutional models. The benefits to your services from including an individual with these skills as part of a design team at every step of development are well worth the investment.

So how do you get started? If you don’t have the luxury of recruiting an individual that already has these skills for a full-time position (there are a number of these individuals already working in libraries!), practical guides such as the 2002 CLIR publication Usage and Usability Assessment: Library Practices and Concerns and the 2001 LITA guide, Usability Assessment of Library-Related Web Sites can be a starting point. (I hear a revision of the latter guide may be in the works.) I’d also suggest reading reports from usability studies (here’s a sample written by a dedicated usability librarian at my institution). As libraries increasingly embrace user-centered design techniques, programs on these topics are becoming more common at library conferences, including programs at ALA sponsored by LITA. In addition, case studies describing user-centered design practices in successful projects are appearing more frequently in the library literature. Specialized conferences such as CHI and training sessions such as those from Edward Tufte are more directed experiences for institutions with larger budgets.

No matter how it is achieved, dedicating staff to ensure services are designed with user needs as the most important consideration is a key part of a successful project. Don’t just go through the motions of task-based testing because that’s what you think you’re supposed to do – make a commitment to ongoing user-centered design by cultivating this expertise in your institution.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Good roundup of methods. That first one you mention--coming in late with a task-based evaluation--is also called "validation testing."

One warning: sometimes "heuristic evaluation" is a euphemism for "we just looked at it ourselves and decided if it was usable."

Also, I agree with Stephen Krug (author of "Don't Make Me Think") that ANY usability testing (design, whatever...) is better than NO usability testing. You're right that to do it well requires an investment in time and expertise... but I'd hate to see libraries shy from usability completely because the bar seemed too high or they felt they had to have a highly skilled usability expert on board.

I also take issue with your comment that "many" libraries do usability testing (even task-based); I'd say "some" is more realistic--particularly if you broaden the user pool BEYOND LIBRARIANS... but perhaps you were just being kind!

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.