The biggest roadblock to DevOps success? Cultural issues

Puppet's 10th annual State of DevOps report shows that cultural issues, such as unclear responsibilities or lack of support from leadership, are the biggest hurdles for organizations stuck in the middle stage of a devOps evolution.
Written by Stephanie Condon, Senior Writer

DevOps practices have been widely embraced across IT organizations, according to Puppet's annual State of DevOps survey. However, the vast majority of IT teams are stuck in the middle stages of their DevOps evolution. And while many people may be quick to blame problems such as a lack of necessary technical skills or legacy infrastructure, the biggest roadblock turns out to be a collection of cultural issues, the study suggests. That includes problems like a lack of support from leadership or unclear responsibilities among teams. 

According to the report, just 18% of respondents have high-functioning DevOps teams. Just 4% are low functioning, while most -- 78% -- are in the middle.

"One way to sort of think about this is that we see folks at the lowest levels really just doing things the way they've always been done," Puppet field CTO Nigel Kersten said during a roundtable discussion. "They haven't modernized very much. The folks at the middle level have managed to optimize for the team in a couple of different ways, but they haven't optimized... for the larger organization. The people who are sort of at the highest levels of evolution are the ones who've managed to make these sorts of practices invisible -- it's just how you do work, and it's just the way you deliver software."

Now in its 10th year, the State of DevOps report surveyed more than 2,650 IT, development, and information security professionals in March and April of this year. The DevOps model assesses respondents over five stages of DevOps evolution, which are used to categorize respondents into three categories: low, mid-level and high DevOps evolution.

In this year's survey, 83% of IT decision-makers report their organizations are implementing DevOps practices to unlock higher business value through better quality software, faster delivery times, more secure systems and the codification of principles.

Also: The snags holding back DevOps: Culture, delivery and security

While organizations in the middle say they're building a DevOps culture, "there is this still organizational resistance to change that is a real problem," Kersten said. "And people really haven't seen the actual value that they're actually trying to move towards with DevOps."

The most common cultural blockers at the mid-level include a culture that discourages risk (21%), unclear responsibilities (20%), de-prioritizing fast flow optimization (18%) and insufficient feedback loops (17%).


"The teams that have a clear understanding of their responsibilities... and understand [the responsibilities of] adjacent teams and their relationships to each other... they're the folks who are better at DevOps," Kersten said. "It turns out highly functioning organizations are better at stuff."

Among highly-evolved teams, 91% report a clear understanding of their responsibilities to other teams compared to only 46% of low-evolution teams. Meanwhile, 89% of highly evolved teams report members of their own team have clear roles, plans, and goals for their work, compared to just 46% of low-evolution teams.

While 77% of highly evolved teams state that teams adjacent to their own team have a clear understanding of their responsibilities as they relate to their own team, only one third of low-evolution teams claim the same.

The report found that highly-evolved firms were better at leveraging automation and the cloud, though these practices were not synonymous with DevOps. 

Ninety percent of respondents with highly evolved DevOps practices report their team has automated most repetitive tasks, and 97% of respondents with highly evolved DevOps practices agree that automation improves the quality of their work. At the same time, as many as 62% of organizations stuck in mid-evolution report high levels of automation.

Almost all survey respondents said they are using the cloud, but most are using it poorly.  Sixty-five percent of mid-evolution firms report using the cloud, yet only 20% use the cloud to its full potential. High-evolution teams use cloud better with 57% satisfying all five NIST cloud capability metrics compared to only 5% of low-evolution respondents. 

Editorial standards