I give a talk on programmer productivity a few times a year, and measuring it (related to this) is very hard. That said, there are a few things worth keeping an

I give a talk on programmer productivity a few times a year, and measuring it (related to this) is very hard. That said, there are a few things worth keeping an eye on. The biggest is turnover. Are people quitting because of workload or general dissatisfaction? Turnover is crazy expensive (I should know, I'm in staffing.) You have others related to technical debt, which is what you get when people are overworked. - Estimate accuracy (estimate vs. actual). Are you increasingly surprised at how long things take? - average hours per issue (not great, but worth keeping an eye on) This assumes you've got good practices like issue tracking, estimates, etc.