I try to believe in the idea of specifying results and letting people use their own methods. One of the issues I wrestle with is that of office standards and whether they get into describing methods instead of results.
I work for a 50-person architecture firm, and we try to standardize some things on the end-product so it all looks like we work for the same company, things like drawing organization, drafting standards, level of detail... you know, basic QC stuff. For that, there's usually not much argument.
The other thing we try to standardize is organization of projects within the company, things like electronic project file organization, CAD layer usage, etc. The goal here was interchangeability, so that in case of a crunch (or a vacation or a termination), other staff members can fill in and be able to find their way around. Since we do a lot of similar (repetitive) work, we can also use it to copy data recursively from one project to another more easily.
We wrote the standard for "worst case" projects, assuming that smaller/simpler projects would fit within it. To simplify it, we have a browser-based creation system that creates all the empty folders for the project.
I've gotten some resistance from one staff member who says that the second standard set is micromanaging, and as long as she meets her deadlines and is on budget (the results she wants), her filing and organization don't matter. Her electronic filing methods are... creative, and really difficult to navigate (it was one of her staff that brought it to me because they could not figure out where something was when she was out one day)
I made the argument that a desired result is the ability for other staff members to work in her project along side her without tutorials on how she likes to do things (which also requires her to re-organize the folders created by the script and create several new ones). With the average project having nearly 4 gigabytes of data and at least 3 permanent team members and 1 to 2 part timers floating on and off, it's important that everyone be able to find what they need. (And in case anyone wonders, yes the filing standards evolve to keep up with technology and changing needs, usually with an annual update/revision published).
Am I specifying methods too much? Or is compliance with a filing standard something that could be considered "a desired result"?
Another side of me worries about the individual. If we let her get away with something minor, will she keep testing the waters to see what else she can get away with? (yes, my issues with her go beyond the filing, but her argument about methods vs results did make me question that one issue)