Consistency is often lauded as a high ideal to strive towards when designing user interfaces. It makes sense – if you do something once, and then a second time, suddenly that thing can become quite natural and … well … intuitive.
Last month I came across a cautionary tale related to consistency. If you create one way of doing something, there is a strong lure to re-use that way whenever possible. This reduces development costs and design costs – both good things to reduce when it makes sense. However this leads to times where your re-used design may be applied in places where it doesn’t quite fit. In fact it might even be applied in places that result in poor usability. This can be particularly true when you’re near the end of a project and “good enough” has become the best you can expect.
How do you avoid this? Ensure that your design is well tested by real users. It is an old refrain but it is a good one – test, test, test again. If you find that some of the design re-use is resulting in bad experiences, poor performance of the interface, or seems challenging to the user then you have some very strong arguments for updating the design. Without the testing you’re left with a bad feeling and perhaps some alternative designs that will be “looked at in the next release”.