Only 38% of organizations have a clear strategy when implementing DevOps automation.

Automation helps teams improve software quality and reduce costs; However, organizations have only automated 56% of their DevOps lifecycle.

Dynatrace has announced the results of an independent global survey of 450 IT professionals responsible for DevOps and security automation in large organizations. Research shows that organizations’ investments in DevOps automation yield significant benefits, including a 61 percent improvement in software quality, a 57 percent reduction in deployment failures, and a 55 percent reduction in IT costs.

However, in most organizations, DevOps automation practices are still in the early stages of maturity. The lack of a clearly defined DevOps automation strategy, the prevalence of toolchain complexity, and the challenges of analyzing surveillance and security data prevent them from realizing the full impact of their investments. This research highlights the need for data-driven and artificial intelligence (AI)-driven automation techniques that enable organizations to better respond to business needs.

He Dynatrace DevOps Automation Report 2023 available for download. Additionally, organizations are encouraged to complete the Dynatrace DevOps Automation Assessment to understand the maturity of their practices.

Some of the report’s findings include:

• Over the next 12 months, organizations will invest in DevOps automation to support security and compliance management (55%), infrastructure provisioning and management (52%), and performance optimization (51%).
• However, only 38% of organizations have a clearly defined DevOps automation strategy to support these investments.
• On average, organizations have managed to automate just over half (56%) of their end-to-end DevOps lifecycle.
• The average organization uses more than seven different DevOps automation tools.
• The biggest barriers preventing organizations from automating new DevOps use cases are security concerns (54%), data challenges (54%) and toolchain complexity (53%).

“As more organizations adopt cloud-based software delivery, DevOps automation has become a strategic imperative,” said Bernd Greifeneder, Technical Director at Dynatrace. “The proliferation of Kubernetes architectures and technology stacks that outstrip human management drives the need for automated orchestration and ecosystem security. Organizations are trying to meet this need by creating and managing automation scripts using a growing set of open source tools combined with DIY approaches and manual efforts. However, cracks are beginning to show in this fragmented approach. Teams are entrenched in data silos, silos of automation, and reactive and manual operations and security efforts. Teams urgently need a unified approach to AI-powered DevOps automation, otherwise it will be impossible to accelerate innovation while maintaining software quality and security.”

Additional findings from the study include:

• 71% of organizations use insights and surveillance data to make automation decisions and improve DevOps workflows.
• However, 85% of organizations face challenges using security and surveillance data to automate DevOps.
• The top three challenges organizations face include inaccessible data (51%), siled data (43%) and the need to transfer data across multiple systems for analysis (41%).
• 54% of organizations are investing in platforms to facilitate tool integration and collaboration between teams involved in automation projects.
• 59% of organizations expect large language models (LLMs) such as ChatGPT and Bard to have a significant impact on their DevOps automation capabilities, with the top three benefits including increased productivity and reduced manual effort (57%), improved development, security and operations . collaboration (56%) and the ability for teams to automatically generate code (48%).

“Data-driven automation is the key to unlocking innovation and meeting customer expectations in the cloud era.” Greifeneder continued. “This requires a platform that can handle the enormous volume and variety of data generated by cloud stacks, and that leverages artificial intelligence to provide accurate and actionable insights for DevOps automation. Unlike traditional AI methods, which have limited scope and applicability, platforms that combine predictive, causal, and generative methods can excel in specific capabilities to address various DevOps automation use cases. “This way, teams maximize the value of their data, eliminate data silos, and can automate DevOps processes with confidence.”

The report is based on responses from a global survey of 450 IT professionals responsible for DevOps and security automation in large organizations, including 150 in the US, 150 in EMEA and 150 in Asia Pacific. The study was conducted by Coleman Parks on behalf of Dynatrace.