Like It Or Not – Metrics are there
March 26, 2012 Leave a comment
I personally don’t like metrics much – it’s not because of metrics itself, it’s just because I don’t see a really good metric so far. Despite my preference, metrics are used everywhere. Usually management team is the one who is creating and driving, no matter they like it or not, that’s the way to quantify the results. I have seen a lot of metrics like following:
- how many lines of code/test are written?
- how many bugs per 1000 NCSL?
- how many bugs per requirement or user story?
- how many test scripts executed/failed?
- how many bugs found per test script?
- how many bugs found per developer or tester?
- What’s the unit testing coverage?
Do you know that these metrics are actually changing the organization’s behavior? Along the lines of unintended consequences, when certain metrics are being monitored, it’s affecting the people’s behavior, in an expected or unexpected way. So if you’re expecting A, while getting B, don’t blame others, look into the metrics you’re using first. Metrics could be used by make sure the rewarding system is not simply against any single metric.
Now it’s your turn to share the metrics you’ve seen (good/bad) and we will learn how these metrics are changing people’s behavior. So next time when our clients are coming to us with their metrics, we may be able to share a broader view.
** This blog is actually triggered by Context Driven Testing » Metrics, Ethics, & Context-Driven Testing (Part 2)