Definition: Set of four key metrics measuring engineering team software delivery performance: deployment frequency, lead time, MTTR, and change failure rate.
— Source: NERVICO, Product Development Consultancy
What are DORA Metrics
DORA metrics (DevOps Research and Assessment) are a set of four key metrics identified by Google’s DORA research team to measure software engineering team performance. The four metrics are: deployment frequency (how often code is deployed to production), lead time for changes (time from commit to code running in production), time to restore service (MTTR, how long it takes to restore service after an incident), and change failure rate (percentage of deployments causing a production failure).
How they work
The metrics are classified into four performance levels: elite, high, medium, and low. Elite teams deploy multiple times per day, have lead times under one hour, recover from failures in under one hour, and their failure rate is below 5%. Low-performing teams deploy less than once a month, their lead time exceeds six months, they take over a week to recover, and more than 46% of their deployments cause failures. The metrics are collected automatically through CI/CD tools, monitoring systems, and incident records.
Why they matter
DORA metrics provide empirical evidence that speed and stability are not at odds: elite teams are both faster and more reliable than low-performing teams. This counters the common belief that moving fast implies more errors. Additionally, they offer an objective benchmark for teams to evaluate their performance, identify improvement areas, and measure the impact of changes to their engineering practices.
Practical example
A development team measures their DORA metrics and discovers they deploy once per week (medium performance), but their lead time is 10 days (low performance) because the code review process takes an average of 5 days. They implement a team agreement to review pull requests within 4 hours and adopt trunk-based development with feature flags. Three months later, they deploy twice daily, lead time drops to 3 hours, MTTR decreases from 4 hours to 30 minutes, and the failure rate remains stable at 8%.