DORA Metrics

The DORA metrics are enabled after CI/CD (this will generate the Deployment Frequency and Delivery Lead Time metrics) is integrated and Incident Management is integrated (this will genearte the Change Failure Rate and Mean Time to Restore)

DORA metrics provide a standardized way to measure software delivery performance across engineering teams. By tracking deployment frequency, lead time, change failure rate, and time to restore, teams gain a consistent and objective understanding of how code moves from development to production and how reliably systems operate. These metrics help establish a shared baseline, making it easier to compare engineering health over time and identify systemic bottlenecks in the delivery process.

For engineering organizations, DORA metrics enable data-informed decisions and continuous improvement by aligning teams on practices that improve both delivery speed and system stability. Over time, this results in more predictable releases, faster incident recovery, and greater confidence among engineers and stakeholders through clear, trusted performance indicators.

To get started on seeing the DORA metrics, you will have to integrate the CI/CD and Incident Management tool - to get visibility into the respective metrics.

Deployment Frequency

The number of times that code changes are successfully deployed to production per day. This is calculated as an aggregate of all deployments in the time period selected.

Delivery Lead Time

The time from first code commit of a PR to successful deployment in production. This is an average of all PRs that have been linked to deployments in the selected time period.

Change Failure Rate

Number of incidents that required code changes / Total number of deployments in the time period.

Incidents typically involve hotfixes, rollbacks or outages.

Once the Incident Management is integrated, we get the number of incidents in a certain time period and out of those only the ones linked with any commits or pull requests will be considered for the calculation of Change Failure Rate

Deployment API is also required to be integrated (i.e. data from CI/CD) so that that acts as the denominator

Mean Time to Restore

The mean time to resolve incidents caused by code in production.

This is measured by incident start date to resolved date.