Change Failure Rate

Definition: Change Failure Rate measures the percentage of deployments to production that result in a degraded service or require immediate remediation (e.g., a hotfix, rollback, or patch). It is a key measure of software quality and deployment stability.

Why It Matters

Change Failure Rate is a critical DORA metric that balances the speed of Deployment Frequency with the stability of the system.

• Measures Quality and Stability: A low CFR is a strong indicator of a healthy, high-quality engineering process. It means changes are being shipped reliably without causing customer-facing issues.

• Builds Confidence: When CFR is low, teams have more confidence to deploy frequently. This creates a virtuous cycle where smaller, more frequent deployments further reduce the risk of failure.

• Highlights Process Weaknesses: A high CFR often points to underlying problems such as inadequate testing, a complex release process, or insufficient code review.

How to Measure It

Change Failure Rate is calculated as a percentage:

CFR = (Number of Failed Deployments / Total Number of Deployments) * 100

A "failure" is typically defined as any change that results in a production outage, a degraded service, or requires an emergency fix.

Interpretation & Benchmarks

• Goal: The primary goal is to achieve the lowest possible Change Failure Rate.

• Balance with Speed: CFR should always be analyzed alongside Deployment Frequency. The goal is to deploy quickly and safely. Elite teams prove that speed and stability are not mutually exclusive.

• Industry Benchmarks (DORA):

  • Elite: 0-15%

  • High: 16-30%

  • Medium: 16-30% (Note: DORA combines High and Medium for this metric)

  • Low: 46-60%

Related Metrics

• Deployment Frequency

• Lead Time for Changes

• Rework Rate