So I was reviewing a design earlier for a reporting project for something that really boils down to glorified turnaround time tracking. Currently we don't capture all the time points needed, so the project entails capturing additional data and then summarizing it. Straightforward enough, really, it's just capturing some time points, taking the differences, and calculating a handful of statistics. I was reviewing the design:
* Known Problems: This design cannot account for [condition X]. In this case they can use [workaround Y].
In calling a few customers and chatting, I confirmed my suspicions:
A) [Condition X] occurs about 95% of the time.
B) [Workaround Y] adds significant extra work to record the data.
C) Use of [workaround Y] screws up the initial time point, which makes most statistics (including both of the metrics that were the raison d'etre for this project) completely meaningless.
Who wouldn't love this feature? 19 times out of 20, they get to do extra work, and they generate useless metrics to boot. On the plus side, their meaningless data is presented in a very attractive report.