by Brian Goodman and Anand Swaminathan
The end of an era is coming for tech-enabled organizations. The workers who set up the infrastructure that makes the operations of most organizations possible—infrastructure such as core systems, software engineering, and physical hardware related to the internet itself—are set to retire in the next two decades or so. If they don’t pass along their expertise, we estimate that by 2050, few people will know how core systems work.
This is a by-product of the mixed blessing of tools that make development more accessible to a wider swath of people. Consider that the worldwide number of developers is projected to reach 28.7 million by 2024, 25 percent higher than it was in 2018.1 Many developers who started their careers after 2010 received training—reinforced by platforms—that de-emphasized knowledge of foundational code and system-level knowledge and thinking. In this context, the coming retirements of technical professionals who have a foundational knowledge of core systems and software engineering will be a significant brain drain.
The challenge for tech leaders to design and maintain resilient software organizations while professionals with basic knowledge of key systems retire is expansive. To help address this challenge, first, leaders would ideally have ambitious goals against which to measure their organizations’ performance on product centricity, agile, DevSecOps, and site reliability engineering (SRE). Second, they would build and revitalize a culture of software engineering that values a mix of deep skills, experience, and the progressive adoption of innovative, effective practices. Finally, they would create developer experience initiatives that map onto the developer journeys to feed the kind of software engineering culture the organization is building. To be sure, these actions are not a comprehensive solution, but they are crucial investments for any technology organization.
Eroding skills and brain drain
Relatively few people have the skills to work with the basic infrastructure that undergirds the technologies organizations rely on (exhibit).
The rise of virtual infrastructure, which is now ubiquitous, has removed the burden of knowing what basic infrastructure requires and has shifted knowledge and control of systems from individual organizations’ technology professionals to third-party hyperscalers. Similarly, platforms that facilitate the building of core runtime components and apps allow less-experienced developers to build solutions with less knowledge and a more surface-level (or even no) understanding of the underlying systems. Tech professionals working in today’s technology environment are therefore less likely to be in touch with the resources required to build and maintain their systems and infrastructure. That knowledge now resides with vendors whose offerings map virtual units to underlying hardware.
In plainer language, consider the analogy of an assembly line of complex machinery. Workers who work on individual components may know their parts of the production process, even intimately. However, few people know how the entire system fits together. This is the approximate position in which tech-enabled organizations across industries find themselves.
Technological trends also dovetail with the steady attrition of talent with a deep, foundational knowledge of core infrastructure. We predict that by 2050, the knowledge of how core systems work will be concentrated among a small number of developers, and most of the industry will rely on third-party providers for this expertise.
Creating stronger software organizations
Individual organizations may not be able to entirely offset the effects of these structural trends, but tech leaders can do three things to strengthen their software organizations. They can use their current performance to establish a baseline and set ambitious goals, build a culture of excellence in software engineering, and design developer experience initiatives that would support their goals.
Find a baseline
Many organizations have parallel technology initiatives. Without coordinating these efforts, it can be difficult to know which can deliver near-term impact while supporting long-term objectives. Establishing baselines and benchmarks for practices that deliver on the dimensions of speed, quality, resilience, security, experience, and productivity can help decision makers understand the approaches that create the most value.
For instance, a large software company that had adopted cloud, DevSecOps, and SRE ran diagnostics to ensure its priorities were sound. Significantly, its benchmarking exercises uncovered new opportunities for the organization.
Build an engineering culture that values a diverse mix of skills
A high-functioning engineering culture should value a diverse mix of deep skills and experience and progressively adopt emerging practices that have been proven to boost organizational effectiveness, such as product centricity, agile, DevSecOps, and cloud.
Crucially, the organization and its culture should emphasize and reward hands-on experience. Systems engineers who can author programming languages and frameworks thanks to their knowledge of the fundamentals—a bare-metal understanding—are uniquely valuable to organizations. Strong technical leadership, clear career paths, and continuous learning for developers can all help retain the right talent and develop better up-and-coming leaders.2
Critically, technology organizations can harness technical leaders’ expertise in software and systems to mentor, coach, and teach colleagues who are earlier in their careers. Over time, the result could be stronger and more diversified talent, higher productivity, and a better workplace.
To retain these developers, leaders could design career paths as part of the change management process when introducing new technologies and approaches such as DevSecOps and cloud. Indeed, any organizational change can present opportunities for career development and clarifying individuals’ contributions to the team. One North American bank rolling out cloud, DevSecOps, and SRE spent as much time planning the change management program as it did on the technical execution. This change management program included continuous-communication plans, education, and coaching programs. The changes will take years to be fully implemented and take effect, but the initial planning has helped the bank organize its approach to change.
Create developer experiences that dovetail with organizational goals
Developer experience is an evergreen pursuit. Boosting engineers’ effectiveness, by definition, can tap into their peak potential earlier in their careers, all in service of technology organizations’ long-term goals and developers’ career growth.
A curated developer experience is about reducing the effort it takes a developer to become productive in a new environment. The learning experience should include more than just classroom time and incorporate a continuous focus on tools and processes that keep talent focused on making distinctive contributions in areas such as SRE, systems thinking, DevSecOps, and product. One North American bank launched an effort focused in part on developer experience. It is in the process of redesigning the onboarding experience to shorten new developers’ time to productivity from weeks to 24 hours. To accomplish this, the organization streamlined the continuous integration/continuous delivery (CI/CD) pipeline and redefined the most important elements of an engineer’s first 30 days to optimize growth.
Organizations will continue to contend with the dual trends of increasing abstraction from fundamental knowledge of tech infrastructure and the loss of senior talent with that knowledge. For now, optimizing their organizations for strength can serve as a bulwark.
Brian Goodman is an associate partner in McKinsey’s New York office, and Anand Swaminathan is a senior partner in the Bay Area office.
1 “Number of software developers worldwide in 2018 to 2024,” Statista, March 6, 2023.
2 For more on recruiting and managing technology talent, see Sven Blumberg, Ranja Reda Kouba, Suman Thareja, and Anna Wiesinger, “Tech talent tectonics: Ten new realities for finding, keeping, and developing talent,” McKinsey, April 14, 2022; and Kathryn Kuhn, Eric Lamarre, Chris Perkins, and Suman Thareja, “Mining for tech-talent gold: Seven ways to find and keep diverse talent,” McKinsey, September 27, 2022.