Tuesday, May 02, 2006

All roads lead to Rome, but you might still miss the approach road

On the importance of your user's mental model of your system...

"When going from basic conceptual design of a system to more detailed design, one of the questions is where to locate functionality, e.g. calling certain helper applications and invoking specific operations. (Note: I am not talking about basic “toolbar functionality” such as Saving here, which should usually be available system wide.) When discussing those issues, sometimes people tend to indulge in a kind of “flexibility rush”, meaning that they start placing one function in multiple locations all over the interface: “When users are working with this part of the interface, they might want to trigger the function, so lets put the button right here.” The cause is a noble one, namely supporting users in getting their work done efficiently, but the best solution might not be that straightforward.

First of all, one should ask, how likely it is, that users might need access to a certain function while performing a given task. Especially subject matter experts may always be able to produce an example that seems to prove the point, but it should be investigated whether such a case arising is more likely than hell freezing over. If not, it’s a good choice to give other insights on system usage higher priority.

Second, even if users might want to access a certain function in the context of a task, it might be worth asking whether it would kill them to perform one additional click to access it because it is not located directly on the interface for their current task. This may seem weird (“Why don’t we give users what they need – everywhere they may need it?”) but it makes sense when one starts to think of users’ mental models. It is not only important for users to be provided with required functionality – it is equally important for them to know where to find it and to come up with a solid mental model of the system structure; the more complex the system, the more important. Stuffing a function in all imaginable places of a system may impair the structure that is communicated to users, because the individual parts of the system become less distinct. (Not to speak of the “button pollution” that can arise.) In the worst case, users don’t know “where to go” to perform a certain task because the individual parts and functional groups of the system are not clearly separated in their mental model. Then one might need to put all functionality on one screen to help them out… (Which, of course, is just not feasible or desirable in real-life interface design.)"   continued ...   (Via Another Useful Blog)


Post a Comment

<< Home

<< Home