All developer productivity stats are flawed, and can be gamed
All developer productivity stats are flawed, and can be gamed.
But, like it or not, I’ve seen managers quietly monitor these 5 stats:
Number of commits
Number of PRs merged
Number of PRs reviewed
Number of tickets closed
Number of story points completed
These stats tell part of the story. But they ignore many relevant concerns:
Is their code quality high?
Do they help others?
Do they have a good attitude?
Do they contribute to conversations?
Are they proactive and self motivated?
Are they reliable, trustworthy, and consistent?
Objective measures are flawed, yet useful.
Objective measures:
✅ Avoid personal bias
✅ Compare everyone using the same stat
🚫 Can be gamed
🚫 May measure the wrong things
Managers who use objective measures aren’t “bad”. Managers who use *only* objective measures are.
Subjective measures are flawed too. They can lead to a manager giving a bonus to the developer who is nicest to the boss, or best at playing political games.
There’s no perfect performance measurement.
So using a mix of objective stats and subjective evaluations is useful.