DORA Metrics

Rating Thresholds

Each metric is rated on a four-level scale based on the thresholds established by the DORA research program. These defaults can be customised in Settings → Rating Thresholds.

Deployment Frequency

What it measures: How often your team ships code to production. Higher deployment frequency is associated with smaller, safer changes and faster feedback loops.

How CI/CD Watch calculates it: Counts the number of successful pipeline runs on your configured deployment branch (default: main or master) within the selected time window, then divides by the number of days in that period to produce a deployments-per-day rate.

Data source: Successful pipeline runs on deployment branches, filtered by the time window you select (7, 30, or 90 days).

Lead Time for Changes

What it measures: The time it takes for a commit to reach production. This captures your entire pipeline duration plus any queuing or waiting time.

How CI/CD Watch calculates it: For each deployment (successful pipeline run on the deployment branch), calculates the time between the commit timestamp and the run completion time. Reports the median across all deployments in the period. Deployments without a commit timestamp are excluded from the calculation.

Data source: Commit timestamps and pipeline run completion times on deployment branches. Your CI provider supplies the commit timestamp; CI/CD Watch records when the run finishes.

Change Failure Rate

What it measures: The percentage of deployments that result in a failure. A lower rate means your changes are more reliable.

How CI/CD Watch calculates it: Takes all completed pipeline runs (succeeded or failed) on branches that have deployments. The rate is the number of failed runs divided by the total completed runs. Runs with other statuses (running, cancelled, etc.) are excluded.

Data source: All completed pipeline runs on deployment branches within the selected time window.

Mean Time to Recovery (MTTR)

What it measures: How quickly your team recovers from failures. Shorter recovery times mean less impact on users and faster incident resolution.

How CI/CD Watch calculates it:Groups pipeline runs by repository and branch, then walks through runs chronologically looking for failure-to-success sequences. For each sequence, recovery time is measured from the first failure's start time to the next success's completion time. Reports the median recovery time across all such sequences.

Data source: All pipeline run statuses and timestamps on deployment branches, ordered chronologically per repository and branch.

Metric Trends

The trends page (under Metrics → Trends) shows charts for all four DORA metrics over time. For 7-day and 30-day windows, data points are shown daily. For 90-day windows, data is aggregated weekly to keep charts readable.

Use trends to track improvement over time, spot regressions early, and validate that process changes are having the intended effect on your delivery performance.

Deployment Detection

CI/CD Watch determines what counts as a "deployment" based on branch matching. By default, pipeline runs on main or master are treated as deployments.

If your team uses a different branch for production deployments (e.g. production or release/*), you can configure the deployment branch in your connection settings. Only pipeline runs matching the configured deployment branch are used for DORA metric calculations. You can also fine-tune how lead time and MTTR are measured in Settings → Metrics Configuration.