What DORA Is and Where It Comes From
The DevOps Research and Assessment (DORA) team — originally an independent research group, now part of Google — spent years surveying tens of thousands of technology professionals worldwide. Their goal was to find out what actually separates high-performing engineering organizations from low-performing ones.
Their most important finding: high performers do not trade speed for stability. They achieve both simultaneously. A team that deploys frequently is not reckless — it is disciplined. Frequent, small deployments are inherently safer than infrequent, large ones.
DORA distilled software delivery performance into four measurable metrics, often called "The Four Keys."
The Four Keys
Velocity Metrics (Speed):
-
Deployment Frequency: How often does your organization successfully deploy code to production? High performers deploy on-demand, multiple times per day. Low performers deploy once a month or less. This metric captures how often value reaches users.
-
Lead Time for Changes: How long does it take from the moment a developer commits code to that code running successfully in production? This includes code review time, CI pipeline duration, approval gates, and deployment time. High performers measure this in hours; low performers measure it in months.
Stability Metrics (Reliability):
-
Change Failure Rate: What percentage of production deployments cause a degraded service — requiring a hotfix, rollback, or incident response? High performers keep this below 15%. This is not a measure of developer skill; it is a measure of how well the system catches problems before they reach users.
-
Time to Restore Service (MTTR — Mean Time To Recovery): When a production failure does occur, how long does it take to restore normal service? High performers restore service in under one hour. This measures the resilience of your system and the effectiveness of your incident response process.
The Platform's Role in DORA
A well-built platform should move all four metrics in the right direction. Automated CI/CD pipelines increase deployment frequency and reduce lead time. Automated testing, canary deployments, and feature flags lower the change failure rate. Runbooks, automated rollbacks, and observability tooling reduce MTTR.
When you present DORA metrics to leadership, you are not showing off engineering vanity stats — you are showing the business impact of the platform investment in terms of speed and reliability.
Interviewers often ask: "How do you measure the success of a platform team?" A weak answer lists tools. A strong answer names DORA metrics, explains what each measures, and then describes how the platform specifically influences each one — for example, "our automated rollback feature directly reduces MTTR by eliminating the manual steps in our incident runbook."
DORA Performance Bands
DORA research categorizes organizations into four performance tiers. Knowing where your organization sits gives you a baseline to improve from:
| Tier | Deployment Frequency | Lead Time | Change Failure Rate | MTTR |
|---|---|---|---|---|
| Elite | Multiple times/day | < 1 hour | 0–15% | < 1 hour |
| High | Daily to weekly | 1 day – 1 week | 0–15% | < 1 day |
| Medium | Weekly to monthly | 1 week – 1 month | 0–15% | 1 day – 1 week |
| Low | Monthly or less | 1–6 months | 16–30% | 1 week – 1 month |
A platform team's job is to help the organization move up this table — ideally from Medium to High, or High to Elite.