Release Management Best Practices

Overview

Metric	Definition	Measures
Deployment Frequency	How often Deployments occur for a given Asset/Product/Service	Velocity
Total Release Delivery Time	The aggregate time across all Release Iterations in a Release’s full Lifecycle	Velocity
Change Failure Rate	The percentage of Releases that fail at a Readiness Gate, require Rollback, or trigger an Issue	Stability
Average Release Time (ART)	The average of Total Release Delivery Time across a set of Releases or a time period	Velocity (trend)

Note: “ART” is used both for Average Release Time in this chapter and for Agile Release Train elsewhere in this document — the two are distinct and should not be confused. Metrics should be sourced from data already captured elsewhere (Readiness Gate records, Deployment outcomes, Release Documentation) rather than separately estimated or manually reconstructed.

This pairing of velocity and stability metrics reflects a widely-cited industry insight from the DORA (DevOps Research and Assessment) research program, associated with Nicole Forsgren, Jez Humble, and Gene Kim’s Accelerate and Google’s ongoing DORA research: high-performing organizations tend to improve delivery speed and system stability together, rather than trading one for the other.

## Best Practice

Track all four baseline metrics together, pairing the velocity metrics with the stability metric — never optimize for speed alone.

Benefit(s)

• Prevents a false optimization where velocity improves at the expense of stability; gives leadership a balanced view on the Enterprise Release Dashboard.

Antipattern

Optimizing only for Deployment Frequency or Total Release Delivery Time while ignoring Change Failure Rate — incentivizing teams to ship fast and unsafe.