(Originally published in 2009 and still highly relevant.)
In her EDUI 2009 Workshop, Usability Testing Without the Scary, Dana Chisnell discussed how to best support great site design. User testing is essential, even if you only have time to spend one hour with one person. If one person has a problem, it’s legitimate to assume that others do too, and to make design changes based on that data. It’s also important to have management buy-in to doing testing–they must be able to see the benefits of doing the research and implementing what is learned from the tests. Multi-disciplinary teams also support implementation of great design, provided they speak one anothers’ languages and understand one anothers’ skills. Finally, it’s essential to be willing to learn as you go and make changes accordingly. (Note that these criteria are completely in alignment with Jared Spool’s research showing that vision, feedback and culture are the most important elements of developing great design.)
Great designs are :
- Useful–What’s valuable to your user about having this information or this tool.
- Efficient–Users can accomplish their goal quickly (that means you have to know what your users’ goals are).
- Learnable–It should be easy for users to apply what they’ve learned about how to use other sites to using your site (your site should conform to usabilities conventions as much as possible).
- Satisfying–Your site should be one that people are happy to use, one that seems to be easy and doesn’t take too much time.
- Accessible–Implementing accessibility guidelines actually makes your site easier for everyone to use.
You should identify design problems that lead to misinformation, incomplete transactions, and/or necessitate support from administrators or staff.
What kind of assessment to use to answer fundamental site design questions
Dana segments analysis into three areas and defines certain types of assessments to use for each area.
What should the site design do? When you want to answer this question, do usability testing, conduct focus groups, have users participate in the design process, do surveys, conduct heuristic evaluations and set benchmarks.
How should the site design work? To answer this question, use participatory design, paper prototyping, walk-throughs, usability testing and heuristic evaluation.
Does the site design do what we want it to do? Answer this via usability testing, heuristic evaluation, follow-up studies and comparison to benchmarks.
Next: Dana’s pointers on putting together a test plan