DORA's software delivery performance model has evolved from the historical Four Keys to five metrics: Change Lead Time, Deployment Frequency, Failed Deployment Recovery Time, Change Fail Rate, and Deployment Rework Rate. Oobeya connects DORA measurement with delivery flow, quality, workflow, and engineering context so teams can improve with confidence.
CURRENT DORA SOFTWARE DELIVERY METRICS
How long does it take for a committed change to reach production?
Tracks the time between a code commit and successful production deployment. Shorter change lead time usually means faster feedback, smaller batches, and less waiting across review, testing, approval, and release stages.
How often does your organization deploy code to production?
Measures the cadence of production releases. High-performing teams deploy on demand, multiple times per day. This metric reflects your CI/CD maturity and ability to ship small, safe changes continuously.
How long does it take to recover from a deployment that fails?
Measures recovery time after a failed deployment requires immediate intervention. This is the current DORA framing for recovery performance and keeps the focus on deployment-related failures.
What percentage of deployments cause a failure in production?
The ratio of failed deployments to total deployments. This metric reveals the quality and reliability of your release process. Lower rates indicate better testing, review, and deployment practices.
What share of deployments are unplanned rework after production incidents?
Measures the ratio of unplanned deployments that happen because of a production incident. Oobeya content establishes rework, defect, quality, and workflow context; confirm your connected source systems before treating Deployment Rework Rate as an available DORA calculation.
Use this metric with incident, deployment, defect, and workflow data so rework is not confused with planned follow-up delivery.
A practical guide for CTOs, VPs, engineering managers, and platform teams who want to connect DORA metrics with flow, quality, and planning data.
Field guide
DORA + flow + quality context
DORA benchmark language has evolved over time. Use performance levels as directional context, then validate metric definitions and current DORA guidance before comparing teams.
| LEVEL | DEPLOY FREQ. | LEAD TIME | CHANGE FAILURE | RECOVERY TIME |
|---|---|---|---|---|
| Elite | On demand | < 1 hour | < 5% | < 1 hour |
| High | Weekly–Monthly | 1 day–1 week | 5%–15% | < 1 day |
| Medium | Monthly–Biannually | 1 week–1 month | 15%–30% | < 1 week |
| Low | < Once/6 months | 1–6 months | 46%–60% | 1 week–1 month |
Source and current guidance: dora.dev
After more than a decade of research, DORA's software delivery performance metrics remain a practical way to understand throughput, stability, and improvement patterns in software delivery.
Replace gut feelings with a research-backed delivery performance model and consistent source-system definitions.
Identify bottlenecks across your delivery pipeline and track improvement over time with clear benchmarks.
Use DORA to review throughput and instability together so speed improvements do not hide reliability risk.
As AI coding assistants accelerate output, DORA metrics become essential to measure real delivery impact.
Oobeya connects DORA metrics with repositories, CI/CD pipelines, incident signals, workflow data, and quality context so teams can interpret delivery performance without rebuilding reports by hand.
DORA FAQ
Clear answers for engineering leaders researching DORA definitions, measurement accuracy, business value, and how Oobeya turns the framework into a more actionable operating view.
The current DORA software delivery performance model includes five metrics: Change Lead Time, Deployment Frequency, Failed Deployment Recovery Time, Change Fail Rate, and Deployment Rework Rate. Historically, many teams knew DORA as the Four Keys.
You are safe with us
