Using leading indicators with DORA to build high performing teams
DORA metrics measure development outcomes, but to drive change, we also need leading indicators highlighted in the 2023 State of DevOps report, empowering teams to adjust and improve performance.
Mar 28, 2024 • 3 Minute Read
While DORA metrics measure the outcomes of your development process, we need leading indicators in order to drive change.
While we often think of the 4 principle DORA metrics, The 2023 State of DevOps report gives insight into the underlying factors that influence those metrics, like team type, key outcomes, process & technical capabilities, and cultural aspects.
Focusing on these underlying factors, what we call leading indicators, empowers teams and leaders to adjust early and improve their performance in the long run.
Effective engineering leaders understand their organisation's value stream and the pain points felt at the different levels of their organization. By combining DORA metrics with Flow metrics, you can a) identify these pain points earlier and b) see what affect your changes have on your organisation.
In order to do so effectively, you need to find the "roll up" to the DORA metrics and use those to collect feedback on your experiments in a much tighter loop. This allows you to iterate and pivot much more quickly, and therefore achieve the goal of continuous improvement much more reliably.
How do we put all of this into Action?
Remind teams that DORA metrics aren’t a measuring stick, the levels are a result of survey data, so the difference between high and elite could be due to a number of contributing factors.
Remind teams that they shouldn’t be obsessed with being elite, but rather, use these metrics to determine your current baseline and work to improve from there
Dive deeper into Flow to identify the leading indicators that your teams can work on now!
Deployment Frequency:
Look at the Team health insights report and see how Coding Days and Time to Merge are trending. This will give an indication how how quickly you are getting to value for your customers.
Lead Time for Changes:
Look at the Retrospective report, in particular Cycle time, Backflow, and Queue time will give an indication of the bottlenecks in your workflow that are impacting you Mean lead time for Changes, i.e. how long it is taking you to get completed code into production.
Mean time to Recover:
We want to look at how long it takes our teams to get fixes for production issues, into production. The Retrospective report allows us to see where work in progress is piling up in a waiting state.
Change Failure Rate:
We want to catch issues in a timely manner. Looking at metrics on Review Collaboration can provide indications of when we're pushing changes that are too large or when we’re not properly reviewing and testing work
