We are a medium-sized, geographically-dispersed development team using Jira for task management and GIT as our repository. Our developers are generally focused on one primary development project at a time, but their time can often be divided between that project and other, smaller things (sometimes development-oriented and sometimes not). For all development-related work (primary project and otherwise) we track tasks in Jira, but for non-development work we don’t.
Our problem is that we want to track development time against tasks and projects, ideally to their Jira tickets, but we have been unsuccessful more than once at getting this off the ground using Jira’s built-in time tracking. The problems we have run into are that time-tracking against Jira tasks is manual, so without frequent reminders the developers forget or ignore tracking often enough to make the tracking records useless. We tend to get off to a good start, but once it seems like everyone is in the groove and the reminders diminish, the time tracking starts to fall off too.
Our goal is to find a solution that allows us to track development time against tasks without the need for a lot of manual action by the developers, and ideally that is not too big-brothery. We want a good work environment, but we also want to have time-spent data. If there is anyone who has been able to successfully find a solution to this problem, we would love to hear about it.
EDIT...
There are several reasons for wanting to do this, but all revolve around the idea of ensuring our engineers are unblocked and working effectively. An engineer could be getting interruptions from other projects, so their output on their main project is suffering. How do we identify that across several teams of engineers? Is someone's lower than expected productivity because they are not getting sufficient instructions, or because they are getting interruptions, or because they are misunderstanding their tasks?
An early diagnostic of any of these problems would be seeing that an engineer is completing 8 hour tasks in 16 hours on average - or that they are booking 20 hours of project time per week instead of 40 (or whatever are the appropriate numbers for the engineer or the project, of course).
And, no, we're not trying to count keystrokes, or lines of code produced, etc. We're trying to help our engineers be as high-performance as they want to be.
Note, we don't use timesheets now - that's one solution that we've tried in the past but we know from experience the frustrations and limitations of that approach.