Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteThe traditional four DORA metrics are deployment frequency, lead time for changes, change fail rate, and recovery time after a failed deployment. They remain useful shorthand, but they are not the whole current framework: DORA now uses five measures, adding deployment rework rate and refining recovery time. Read them together as a view of delivery throughput and instability—not as a score where faster deployment is automatically better.
What are the four traditional DORA metrics?
DORA’s original “four keys” describe how frequently software reaches production, how quickly changes get there, and how often a production change causes trouble that requires intervention or recovery. The measures apply to changes of any kind, including code and configuration changes.
As an Amazon Associate I earn from qualifying purchases.
| Traditional metric | What it measures | Executive interpretation |
|---|---|---|
| Deployment frequency | How often changes to an application are deployed over a period, or the time between deployments. | One view of delivery throughput. More frequent deployment alone does not establish better performance. |
| Lead time for changes | Time from a code change being committed to version control until it is successfully deployed to production. | How long a committed change takes to reach users through the production path. |
| Change fail rate | The proportion of deployments that require immediate intervention after deployment, such as a rollback or hotfix. | A measure of instability associated with changes reaching production. |
| Recovery time | Time to recover when a production change causes service impairment and requires intervention. | How quickly the service is restored after a failed change; older descriptions often call this MTTR or time to restore service. |
The four are a starting point for a conversation, not a complete diagnosis. They do not, by themselves, explain why delivery is slow or a change failed.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Why does DORA now have five metrics?
DORA’s current software delivery performance guide defines five metrics: change lead time, deployment frequency, failed deployment recovery time, change fail rate, and deployment rework rate. The 2024 addition, deployment rework rate, makes unplanned production bug-fix deployments more visible.
#1 Best Overall
That addition addresses a limitation of relying on change fail rate alone: fail rate is a proxy for rework, but it does not directly capture every unplanned deployment made to fix a user-facing production bug. DORA also narrowed the recovery measure. The relevant recovery is restoration after a production change causes impairment, rather than any service disruption regardless of cause. That distinction avoids treating unrelated events, such as a data-center outage, as a failed software deployment.
How the current five metrics fit together
DORA groups the current measures into throughput and instability. Throughput covers the normal path of delivering changes; instability captures failed changes and unplanned recovery work. A useful executive view considers both sides together.
| Group | Current metric | Question it helps answer |
|---|---|---|
| Throughput | Change lead time | How long does a committed change take to reach production? |
| Throughput | Deployment frequency | How often are changes deployed? |
| Throughput | Failed deployment recovery time | How long does recovery take when a production deployment causes impairment? |
| Instability | Change fail rate | What share of deployments require immediate intervention? |
| Instability | Deployment rework rate | What share of deployments are unplanned and address a user-facing production bug? |
Use the paired view to avoid a misleading “faster is always better” conclusion. A change in deployment frequency or lead time should be read alongside failure, recovery, and rework measures. For example, a frequency increase is not enough on its own to show that delivery improved if instability or unplanned repair work also rises.
How executives should use the metrics
Measure one application or service at a time
DORA says the metrics work for any type of technology an organization delivers, but are best suited to measuring one application or service at a time. Its Quick Check likewise asks about a primary application or service. Avoid combining unrelated services into one number: differences in architecture, release practices, or operational needs can make an aggregate difficult to interpret.
Keep definitions and observation windows consistent
When comparing a service over time or comparing teams, use the same metric definitions and observation window. For failure-related measures, agree on what counts as immediate intervention and which remediation actions qualify. DORA’s examples include a hotfix, rollback, fix forward, or patch. Changing the counting boundary can change the result without any underlying change in delivery performance.
Use trends to prompt investigation, not to declare a cause
The metrics describe outcomes in the delivery and recovery paths; they do not identify the cause of a change. If lead time lengthens, investigate the service’s delivery path. If failed deployment recovery time or rework rises, examine the incidents and remediation work counted. Use the measures to focus discussion and improvement work rather than treating a single metric as a verdict on a team.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How the four keys became five
DORA’s metrics history traces the original four variables to its 2014 study: deployment frequency, lead time for changes, mean time to recover, and change fail rate. Over time, DORA refined recovery to focus on impairment initiated by production changes. In 2024 it added deployment rework rate because change fail rate did not directly measure every unplanned bug-fix deployment. The historical four remain familiar; the current five provide a more explicit view of both normal delivery and repair work.
Quick Recap
Official DORA references
- DORA’s software delivery performance metrics — current definitions, groupings, and guidance on scope.
- A history of DORA’s software delivery metrics — the evolution from four variables to the current five.
- DORA Quick Check — questions framed around a primary application or service.
- 2024 Accelerate State of DevOps Report — the four-key definitions and their application to code and configuration changes.
- DORA Core Research Questions — question framing and remediation examples.
- Value stream mapping for software delivery — the normal delivery path and recovery work.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
