I’ve taken several classes at MIT where we’ve discussed the design process. My personal design process involves establishing the problem, creating prototypes, testing those prototypes, and iterating to improve. I’ve found that designing specifically for my team’s client, Dave, is a lot different than designing for a generic class of users. In many cases, it is easier.
For example, when designing for a class of users, it’s important to make the design general enough that many people can use it. If one user tester has very specific requests, it can be hard to weight how likely other users are to have the same problem without doing a wide reaching survey. Then, that can accidentally prompt respondants to respond in a certain way. Direct observation may be the best way to learn about the needs of a user class, but it is hard to scale to a large number of potential users.
When designing specifically for Dave, however, we can take any suggestions he makes during interviews, observations, and testing to be definitive user needs. For example, if we were designing a TV remote control app to be used by anyone with MS, we wouldn’t know which one of their arms is stronger. We might design a flexible UI which allows the user to position different elements on the screen as works best for them. However, this could lead to a complicated and confusing design. Since we are designing specifically for Dave, we know that he has the best ability to tap the screen on the left side, and thus we can put the most frequently used buttons on the left side. This specificity would not be possible when designing for a class of users, and it is one example of how designing for one user is easier than generalized design.