Details on predefined metrics

[Spencer Robertson]

I am trying to figure out how the predefined metrics are calculated against a Github repo and haven't been able to find any documentation on the subject.

Ive managed to make some best guesses on some of the metrics but would be quite nice to have some extra documentation , specifically on how to set them up, best practises and how they are actually calculated. (I also think this could be helpful for peole making custom metrics, having a baseline on which data is being brought in)

I.E. Deployment time vs Build time:

My guess is this calculates the time for a workflow to finish, and the only difference is if one of the workflows is tagged as "environment: production" then it goes under the Deployment time metric?

Any help, or links to documentation please?

[Alastair Wilkes · Accepted solution]

Hi Spencer, thanks for your question!

Build time and deployment time are derived from build and deployment events sent from GitHub via the GitHub/Compass integration. More details are here:

https://developer.atlassian.com/cloud/compass/integrations/integrate-Compass-with-GitHub/#supported-metrics

You're correct that the integration looks for the environment tag to detect deployment events. The link above goes into this in a bit more detail.

You can see more info about other predefined metrics here: https://developer.atlassian.com/cloud/compass/components/available-predefined-metrics/

For the most part, you shouldn't have to do much to get things set up if you're using the out of the box integration (except check the environment name for deployments, as you pointed out). Let me know if you have any other questions!

Alastair

[Spencer Robertson]

Most of the metrics worked but was just struggling to see how they were generated to "fine tune" them (turns out some of our workflows had incorrect enviroment tags so the metrics were being thrown off)

This page really helped https://developer.atlassian.com/cloud/compass/integrations/integrate-Compass-with-GitHub/#supported-metrics. Thanks for that!