Change Failure Rate
DORA stability metric
Change failure rate is a DORA metric measuring the percentage of deployments to production that result in a degraded service, requiring a hotfix, rollback, or remediation.
Definition
Change failure rate is a DORA metric measuring the percentage of deployments to production that result in a degraded service, requiring a hotfix, rollback, or remediation.
Overview
Change failure rate answers the question: of everything a team ships, how much of it breaks something? It is expressed as a percentage — deployments causing an incident divided by total deployments — and is one of the four DORA metrics used to characterize both delivery speed and software stability together. What counts as a "failure" varies by organization, but common definitions include any deployment that triggers a rollback, a hotfix, a page to on-call, or a documented incident. The metric is deliberately paired with deployment frequency and lead time for changes because those speed metrics can be gamed by shipping carelessly; change failure rate acts as a check, showing whether high velocity is coming at the cost of stability. DORA's research consistently finds that elite performers do not trade stability for speed — they achieve both a high deployment frequency and a low change failure rate (commonly cited in the 0-15% range for elite teams), because practices like automated testing, small batch sizes, and progressive delivery reduce risk on every axis simultaneously. Low performers, by contrast, often show both slow delivery and high failure rates, since infrequent, large releases are inherently harder to reason about and test thoroughly. Teams typically calculate change failure rate from data in their incident management and deployment tools, tagging incidents to the deployments that caused them. Because the definition of "failure" is somewhat subjective, the trend over time within an organization is usually more valuable than comparing raw percentages across companies with different incident-tagging conventions.
Key Concepts
- Percentage of deployments causing a production incident or requiring rollback
- One of the four core DORA metrics
- Acts as a stability check against speed-only metrics
- Elite teams combine high deployment frequency with low failure rates
- Calculated from incident management and deployment tool data
- Definition of 'failure' varies and should be standardized per organization
- Most useful tracked as an internal trend over time
Use Cases
Frequently Asked Questions
From the Blog
What Is a Learning Rate and How to Tune It
The learning rate controls how big a step a model takes when updating its weights during training — the single most important hyperparameter to get right.
Read More Career GrowthCareer Change at 40: Is It Too Late to Switch to Tech?
A career change at 40 into tech is not too late; it simply requires a focused plan that leverages existing experience while filling specific skill gaps. This guide covers how to evaluate the move, choose a direction, and build momentum realistically.
Read More AI & TechnologyWhat Is CTR? Click-Through Rate Explained
CTR, or click-through rate, measures the percentage of people who click a link, ad, or search result out of everyone who saw it. This guide explains how it's calculated, what affects it, and how to improve it.
Read More AI & Technology9 Agent Failure Modes to Test Before You Launch
Agents fail in a small number of recognisable ways, and nearly all of them can be provoked deliberately before a user finds them. This article names nine failure modes, from invented tool calls to stale memory and irreversible actions, and gives a concrete test for each that belongs in a pre-launch suite.
Read More