Even though we are a startup and heavily leaning on result-only work environment model, I noticed it's really hard to do this correctly, we still do have core hours and so many similar stuff and so far didn't manage to implement this successfully.
Is there any good guide on prerequisites for ROWE, or tips and tricks of implementation? Even better can you describe your own experience if you are part of a ROWE company.
I'm talking from the perspective of a software-product company and mostly referring to development team rather than business people.
Its an interesting concept, the best tip I can give is to define what the results are in detail and to break them down into bite sized chunks.
My company breaks a typical design into 4-8 sprints of 2 weeks each, each sprint has roughly 140 tasks which cover between 30 mins and 4 hours.
We started doing this because we wanted the junior developers to know what "done" meant because they would always say "i'm done" and invariably they would be no where near it when we checked.
Each of tasks represents a "thing that needs to be checked off". Each developer can still "do it their way" but it ensures they don't forget anything and is the "sign off" of the results that is important.
We include each known user interface, database table, search and unit test that is indicated by the design. Other come up obviously and they go into the next sprint or if really serious are addressed on the spot.
People by default have to do the minimum hours because the tasks while they have "buffer" built in are still fairly accurate.