What Is the DORA Framework?
DORA (DevOps Research and Assessment) identified four metrics that together capture how well a team delivers software, balancing throughput against stability.
The DORA framework came out of years of research into what distinguishes high-performing software teams. Its lasting contribution is a small, balanced set of metrics that resist gaming: two measure speed, two measure stability, and you cannot improve one pair by sacrificing the other for long.
The four key metrics
- Deployment frequency: how often you ship to production.
- Lead time for changes: how long from commit to running in production.
- Change failure rate: the share of deployments that cause a failure.
- Time to restore service: how quickly you recover from a failure.
Why these four together
The metrics are paired deliberately. Deployment frequency and lead time measure throughput; change failure rate and time to restore measure stability. Tracking all four prevents the classic trap of going faster by cutting corners, because the stability metrics would immediately degrade.
Speed and stability rise together
A central DORA finding is that the best teams are not forced to choose between speed and stability -- they achieve both. Small, frequent, automated changes are easier to test, review, and roll back, so moving faster actually tends to make systems safer rather than riskier.
Using DORA to improve
The metrics are most useful as a compass for improvement, not a leaderboard. A team measures its current numbers, identifies the weakest one, and invests in the practices that move it -- better test automation, smaller batches, faster pipelines, or stronger observability.
Beyond the four metrics
Later research added a fifth dimension around reliability and operational performance, and emphasized that culture and developer experience underpin the numbers. The four metrics remain the well-known core, but they are best read as symptoms of deeper organizational health.
Key takeaways
- DORA defines four balanced metrics: two for speed, two for stability.
- They resist gaming because improving throughput at the cost of stability shows up immediately.
- High performers achieve speed and stability together, not as a trade-off.