Benchmark Your DevOps Performance with the 2019 Accelerate Report

Why DevOps?

DevOps is most valuable for organizations whose mission depends heavily on IT. When IT is a strategic enabler, improving software delivery speed and reliability directly supports commercial goals (profitability, market share, customer satisfaction) and non‑commercial goals (operational efficiency, mission achievement).

Four Golden Metrics

The 2019 Accelerate (DORA) survey clusters organizations by two throughput metrics and two stability metrics. These four metrics are the basis for the “DevOps performance bubble chart”.

Throughput – Deployment Frequency : How often code is released to production.

Throughput – Lead Time for Changes : Time from code commit to a successful production run.

Stability – Mean Time to Restore (MTTR) : Time required to recover from a service incident.

Stability – Change Failure Rate : Percentage of releases that cause service degradation or outage.

How to Benchmark Your Organization

Google provides a public DORA‑based performance database at https://beta.devops-research.com/performance.html. By answering five questions (deployment frequency, lead time for changes, MTTR, change failure rate, and industry sector) the tool generates a personalized benchmark report that places your organization into one of four performance tiers (Elite, High‑performing, Medium, Low) and shows both overall‑industry and sector‑specific standings.

Service‑Operations Efficiency (SRE) Metric

The 2019 report introduces a new service‑operations efficiency metric that complements the four golden metrics. While the four metrics focus on software delivery, this additional metric captures reliability and availability from an SRE (Site Reliability Engineering) perspective, completing a full Software Delivery and Operations (SDO) feedback loop.

Practical Steps for Continuous Measurement

Collect the four DORA metrics via the standard questionnaire (questions 1‑5 of the 2019 Accelerate survey).

Automate data collection where possible (e.g., CI/CD pipelines for deployment frequency, version control timestamps for lead time, incident management tools for MTTR, and change logs for failure rate).

Feed the metrics into the Google minimal‑ist measurement tool or an internal dashboard to obtain real‑time values.

Periodically run the public benchmark questionnaire to see how your organization’s metrics compare to industry averages.

Incorporate the SRE‑based service‑operations efficiency metric (e.g., availability, error budget consumption) to close the loop between delivery and operations.

Conclusion

Mastering the four golden metrics, continuously measuring them with automated tooling, and extending the view to include the service‑operations efficiency metric enables teams to benchmark performance, identify improvement areas, and sustain accelerated DevOps growth.