Survey Shows Increased Reliance on DORA Metrics
A new survey found that just under a quarter (23%) of respondents now track the four DevOps metrics defined by the DevOps Research and Evaluation Team (DORA), with another 17% in follows three.
The Developer Experience Status Survey surveyed 129 IT professionals who play a role in software development and was conducted by LeanIX, a platform provider for managing enterprise application creation.
Dominick Rose, vice president of product management and strategy at LeanIX, said the ability to track DORA metrics such as deployment frequency (77%), failure rate (73%), delivery time of changes (63%) and mean time-recovery (MTTR) (55%) are key indicators of DevOps maturity.
DORA metrics, which are defined and maintained by a Google unit, are most often used to drive agile planning (69%) and optimize continuous integration / continuous delivery (CI / CD) platforms (58%). ), according to the survey.
Overall, the survey revealed that just under half of respondents (47%) said their organization had a high level of DevOps maturity, defined as having adopted three or more DevOps working methods. These working methods are: Being flexible to changes in customer needs; have implemented a CI / CD platform; all engineers build, ship, and own their own code; teams are organized around topologies and each team is free to choose their own stack of technology.
Of course, each individual organization will determine for itself what level of depth of DevOps is required. For example, not all organizations would see the need for teams to be organized around topologies or be free to choose their own stack of technology. In fact, Rose said the survey made it clear that larger enterprise IT organizations tend to have a lower overall level of DevOps maturity. One reason for this is that many larger organizations still use legacy processes to create and deploy software, Rose noted.
Most developers are also more advanced in terms of continuous integration (IC) than IT operations teams in adopting continuous delivery (CD), Rose added. However, as the number of applications being developed with low-code and no-code tools continues to increase, it’s only a matter of time before application delivery has to be further automated, he said. Achieving this goal will be easier as more artificial intelligence (AI) is incorporated into IT operations, Rose added.
However, the survey made it clear that there is still a long way to go in terms of bridging the gap between DevOps teams and the rest of the business, with only 20% of respondents indicating that they use a management platform. value stream (VSM) to track the efficiency of engineering to link development activities to actual business results.
However, a total of 71% of respondents tracked open attendance tickets as a measure of customer value, followed by 66% who tracked monthly users and 45% who tracked adoption of functions.
It remains to be seen how DevOps will be enacted more across organizations, but as more organizations become aware of their reliance on software, it will become a matter of how well they adopt DevOps best practices as the pace continues. to which applications are being built and deployed. to accelerate.