Use Different Metrics for Dashboard

[PaulDuvall]

Use the following metrics:

High Priority:

• Mean Time to Pipeline Recovery. Divide the current Red Time by Failure Count
• Cycle Time (already defined)

Medium Priority:

• MTTR - Mean Time to Recovery. Get production failures to alert pipeline and track the time to recovery. This has a high impact to the business. Marked as medium priority because of the instrumentation and work required to assess this measure.

Low Priority:

• Average Deployment Frequency - Average # of deployment successes to production per day: SuccessCount divided by the time period (i.e. 30 days)
• Average Commit Frequency - average number of commits per day (or per weekday). This could be correlated to the number of engineers on a project/pipeline to indicate whether small or large batches are being committed at once.
• Trending - Show whether the measure is trending up or down based on the 30 days prior to the current 30-day period (or previous week compared to 30-day period)?

Not sure of the value of SuccessCount and FailureCount as individual measures but would like to learn