DevOps: The key to IT infrastructure agility

by Thomas Delaet and Ling Lau

These days, digital grabs a lot of headlines that trumpet how it’s radically changing customer behaviors. This typically means that IT departments have to deliver new features faster even in the face of more demanding requirements for availability (24/7) and security.

DevOps promises to do exactly that, by fostering a high degree of collaboration across the full IT value chain (from business, over development, operations and IT infrastructure). But there’s a problem.

While many software-development and operations teams have made steps toward DevOps methods, most enterprise IT-infrastructure organizations still work much as they did in the first decade of this century: They use a “plan-build-run” operating model organized by siloed infrastructure components , such as network, storage, and computing. The result? Agile development teams slam into a bottleneck when new or updated applications are ready for hosting on the existing IT infrastructure, where the norm is still highly manual processes based on ticket management.

Digital organizations can clear this bottleneck by extending the DevOps model so that application development, application operations, and IT infrastructure work as one. Based on our experience, the benefits of this move include a 25 to 30 percent increase in capacity creation, a 50 to 75 percent reduction in time to market, and a greater than 50 percent reduction in failure rates.

Here are the six fundamental shifts required to extend the DevOps approach into IT infrastructure:

  1. Work as one team

    Rather than organize themselves by highly specialized functions, IT infrastructure teams need to come together to work as one unit whose top priority is to provide end-to-end value to the customer, not to optimize discrete infrastructure components. By pulling together their aggregate skills—infrastructure automation, information security, IT networking, and others—into one team that eliminates hand-offs, IT can deliver the entire infrastructure solution rapidly, often in the form of an API that others teams can use.

  2. Apply design thinking to IT infrastructure

    Digital infrastructure organizations can help businesses deliver faster solutions with more self-service and an exceptional user experience. That requires a customer-centric approach rooted in a deep understanding of all the actions customers take to accomplish a task. They can gain that understanding the same way application developers do: by creating user personas, mapping journeys to identify pain points and delighters, and running rapid test-and-learn cycles. This approach helps prioritize the initiatives that deliver better and faster customer and business outcomes.

  3. Shift to next-generation technical practices

    Infrastructure teams should consider embracing new practices, such as “infrastructure-as-code,” which allows infrastructure teams to adapt software-development engineering practices and ways of working and cast off the less flexible “hardware” mind-set. IT infrastructure functions that can adopt software best practices—such as extensive test automation, continuous delivery of infrastructure, and a rapid test-and-learn approach to infrastructure management—will also be able to easily maintain and upgrade large-scale automation programs, provision and scale hundreds of infrastructure components without a hitch, and identify and resolve issues proactively and near instantaneously.

  4. Invest in building software engineering talent

    Infrastructure-as-code and building out cloud software platforms at scale have shifted talent demand from system administrators to software engineers skilled at building and managing code. To manage this shift, many leading organizations are investing in in-house immersive boot camps that use real-life work examples to teach their IT infrastructure staff the latest and best software-engineering practices. They are also updating their talent-acquisition strategies and seeking out DevOps and cloud engineers with software-engineering experience that they can apply to new infrastructure-as-code technologies. They’re scouting both traditional and nontraditional sources where top talent convenes, including (to name a few) online communities such as HackerNews and StackOverflow, competitive digital platforms such as TopCoder, Kaggle, and Codility, and meet-up communities.

  5. Overinvest in culture change

    Leading IT-infrastructure organizations aren’t just introducing new tools and processes; they’re overinvesting in culture change. That starts with senior executives clearly and frequently articulating the program vision and modeling the DevOps way of working. At the operational level, in addition to providing teams with DevOps coaches and structured programs to build capabilities, evolving IT infrastructure organizations are forming service teams focused on specific end-to-end customer journeys and updating processes and KPIs to accurately track and measure adoption of the DevOps methodology.

  6. Link incentives to delivery and service goals

    The success of a transformation is measured in improved delivery and service outcomes, so incentives and performance management should emphasize team results, not the achievements of individual functions. To reinforce this linkage, leading businesses are putting in place metrics that balance business value, time-to-market, service availability, quality, and overall employee satisfaction. These measures enable IT infrastructure organizations to align the often-divergent goals and perspectives of developers, operational teams, and business leaders.

DevOps has had great impact for developers and operations teams. Extending that way of working across IT infrastructure and other elements of the business is the next source of value for IT.

Thomas Delaet is an associate partner in McKinsey’s Brussels office, and Ling Lau is a digital VP in our New Jersey office.