Improve engineering teams faster with DORA metrics
Improve engineering teams faster with DORA metrics
We talk with numerous teams that want to improve their engineering performance. Here, we explain how to accelerate your progress using DORA metrics — a set of key performance indicators that can help you measure and optimize your team's software development process. You'll learn practical tips on how to leverage these metrics to achieve faster and more efficient team improvement.
But first, you'll need your team to see the value in DORA metrics.
How do you get your team to adopt DORA metrics?
The success of any new process, framework or tool depends on how it’s adopted, and adoption depends on clear expectations for what problem it solves, who will use it, how, and what value they’ll get out of it.
The good news is, people in various roles — individual contributors, senior software engineers, engineering managers, CTOs — can all improve their teams by tracking DORA metrics and acting on the data. It’s a matter of knowing what’s possible with the four DORA metrics and what they can do for you and your team.
Check out these real-world use cases of DORA metrics to spark ideas for how to get your team on board and start improving faster.
Use DORA metrics to test and verify the impact of process changes
This can be particularly helpful for engineering managers, as they’re at the intersection of technology, people and processes.
Let’s say you want to make a process change. You have a prediction about how it will improve your team, but you want data to know if it worked or not. Measuring DORA metrics can provide the connection between predictions and hard data.
For example, maybe an engineering manager wants to change a developer-on-support rotation, predicting that the change will protect the team from context switching and increase deploy frequency.
Once the change is implemented, she’ll be able to review DORA metrics dashboards to see how that process change affects deployment frequency as reflected in trends over time.
DORA metrics for improving developers' flow
Metrics aren’t just for managers. And while developers may be skeptical of metrics, DORA metrics in particular can actually help developers get into and maintain their flow by empowering them to improve their own work.
Take bottlenecks, for example. Removing them is a key problem many engineering teams continuously try to solve in order to improve. The change lead time DORA metric in particular helps zero in on where the bottlenecks are.
Look at the change lead time breakdowns for coding time, review lag time, review time and deploying time to identify which areas need improving and the specific issues that are causing the bottlenecks.
Then, because developers have access to team- and project-level DORA metrics, they can fix the issue right away or offer to help a team member fix it — and drive improvements. And with less time spent figuring out what's causing bottlenecks, developers have a better chance of maintaining their flow, staying focused and avoiding context switches.
DORA metrics encourage small batches for better performance
If you dig into the history of DORA metrics, the research behind them found a correlation that deploying in smaller batches tends to lead to smaller change lead times, smaller failure rates and shorter mean times to recovery.
So, if your team struggles with long PRs, getting anyone to respond to them, or struggles with risky changes so much that management intervenes, DORA metrics and deploying in small batches are natural antidotes to those situations.
Empower teams by measuring DORA metrics
A common theme we hear from customers is, “How do we do more with less?” It’s especially common when teams experience layoffs.
One customer lost its entire QA team and started using DORA metrics to get teams back on track with increasing deployment frequency while maintaining software quality. They’re looking at their large items and breaking them up before deploying to spread out the work.
Another customer empowers teams by giving them easy access to DORA metrics so they have data that shows what work is at risk. They can then be proactive, dive in on their own, make changes without top-down direction, and come to sprints being able to add to the discussion.
A residual benefit of empowering teams with DORA metrics is improved team morale. Seeing their work and the impact of their changes empowers developers to take ownership and make improvements. That’s great for developers and managers alike. And with everyone looking at the same dashboards, there are no surprises. The team looks at team metrics and project metrics together, and makes improvements together.
Power improvements with DORA metrics
If you have a DORA metrics tracking tool, you’ll get the most value from it — and higher adoption rates — by using the data it gives you to fuel improvements.
Once you have DORA metrics tracking in place — i.e., you have a tool that’s connected to your issue system, source control, CI/CD system, etc. — and you’re getting visibility into your team’s performance, take that to the next level and find ways to improve.
Want to improve your deployment frequency? Make a change to your continuous deployment process and then see what effect it had on deployment frequency and quality (as reflected in your change failure rate).
When it comes to getting teams to adopt DORA metrics, think first about what problems and challenges your team faces. If those challenges are around visibility, deployment speed and improvement, or quality, then DORA metrics and a DORA metrics tracking tool could help solve those problems. And something that solves problems is more likely to be used.