This message was deleted.

We think that if infra is requested through APIs, you can measure and track usage.

@Raquel Pau Fernandez Usage only tells if it is being used on not, even in the previous embedded DevOps model, some system or the other was being used to instantiate infra for their projects. I am trying to think along the lines of Developer Productivity metrics. Any you can think of?

Rate of deployment or provisioning?

@Mannar Karyampudi In our model, those APIs should be executed during the deployment process, and therefore you can understand the performance impact in the lead time.

@Mannar Karyampudi • adoption rate (how many developers have moved to IaC, if applicable) • if you have a shared Slack channel for support questions, activity metrics for this channel such as inquiries per day • in the same way you'd measure any product, breaking down the expected flow into a funnel and understanding the latency / success rate between each step. And then capturing the overall lead time of that workflow

Thank you Antoine!

Up

I know it’s not sexy, because it’s so ubiquitous, but the amount of downtime caused by changes in infrastructure seems like a worthwhile metric to me. I feel that

terraform plan

single handedly has dropped this metric by a lot.