Behind every great observability practice is a platform engineering team. Platform engineering can lead to more secure code, more productive developers, and faster time to market. But building these teams — and ensuring their success — isn’t a simple ‘set it and forget it’ process.
In Splunk’s State of Observability 2024, we identified an elite group of respondents who are more successfully adopting platform engineering. Let’s dig into what they’re doing differently.
What platform engineering is — and isn’t
Platform engineering is all the rage (nearly three-quarters of respondents are adopting it). To put it simply, platform engineering enables software engineers to do their jobs better.
Platform engineers are not site reliability engineers (SREs) rebranded. While SREs typically troubleshoot and monitor alerts, platform engineers create a platform for developers, including standards for how developers write, instrument, deploy, operate, and monitor code — integrating observability into software earlier and more efficiently. They also standardize infrastructure, hardware, security practices, and build toolchains and workflows, so that software engineers don’t need to worry about maintaining that infrastructure. Instead, they can focus on what they do best: building reliable, secure software.
Running continuous integration/continuous deployment (CI/CD) software or installing Jenkins doesn’t mean an organization is ‘doing platform engineering.’ While CI/CD is a major part of platform engineering, it’s not everything. Platform engineering is the discipline of building a platform that’s tailored to the business to meet its evolving needs.
Leaders report better efficiency, standardization with platform engineering
Organizations with more advanced observability practices not only adopt platform engineering at a higher rate — 78% extensively employ platform engineering, versus 27% of beginners who do the same — but they can more easily reap the benefits that platform engineering provides.
Splunk’s State of Observability 2024 revealed that organizations with leading observability practices — a subset of respondents that consistently adopted a set of observability best practices, including maintaining better visibility and proactively responding to issues — are more likely to say their platform engineering teams are “very successful” in standardizing across every area we asked about, including documentation and support, instrumentation, and security and compliance controls. They are also more likely to report that platform engineering teams improved IT operations efficiency (63%) and developer productivity (48%).
It tracks that platform engineering teams at these leading organizations have stellar reputations. Leaders more often say that developers view their platform engineering teams as competitive differentiators as opposed to simply valuable service bureaus, mere cost centers or worst of all, innovation hindrances.
Platform success starts with an open-minded culture
Leading observability practices are built within a company culture that is constantly striving to be better. Similarly, a platform team is only as successful as the culture it exists within — especially because platform engineering is inherently centered around changing the way things have historically been done. Some of these changes may be unpopular — for example, roles such as build specialists and release train engineers may be threatened by a platform engineering team that wants to automate releases.
Open-minded leaders, on the other hand, can be catalysts for positive change. Leaders can ensure that platform engineering teams are heard by advocating for their ideas. Designate an executive sponsor to voice their support, so that platform engineers don’t feel unsupported and singled out on their quest for change.
Embracing a culture of change, however, is a gradual process. Leaders must ensure that someone in the organization understands how to properly apply the principles of platform engineering and can help marshal culture change — because it does not happen overnight.
Habits of successful platform engineering teams
Once you feel confident that your platform engineers will thrive within your company culture, consider these practical strategies to take the practice to the next level.
Treat the developer platform as a product. The best platform engineering teams take ownership of their platforms and have pride in what they create. One way to achieve this is to view developers and SREs as customers, and the platform as an internal “product.” This means that platform engineers should seriously consider what their users need and how the platform will satisfy those needs. They should implement practices such as developing availability standards, maintaining a release schedule, and sprint planning.
Close the feedback loop. Leading observability practices measure their performance and don’t settle for mediocrity. DevOps Research and Assessment (DORA) metrics such as change success rate and deployment frequency and compare results to their competitors can validate whether or not the organization’s platform engineering practices are working. This can be done easily using tools like synthetic monitoring.
Equally important is having a feedback loop that drives improvement over time. Encouraging software engineers to think about how to close the loop with each release — such as running synthetic browser workflow tests to validate feature performance and monitor overall customer experience over time—helps ensure continuous improvement and aligns with platform engineering principles.
Collaborate seamlessly. A successful platform engineering team relies on many different stakeholders — including SREs, cloud architects, sysadmins, software engineers, and application security folks — to work together towards a common goal.
If a platform engineering team’s status updates frequently report that other teams are blocking the completion of a goal, then something isn’t working. Leadership should ensure that roles and responsibilities have been properly defined and that all teams embrace open communication.
Read the full report for more insights and recommendations on establishing a successful platform engineering team. The report also covers OpenTelemetry adoption and challenges, and the role of AI in observability.
