Observability Engineering, 2nd Edition

Chapter 24. Systems Thinking for Software Delivery

In the previous chapter, we explored how AI amplifies the existing behaviors of your organization. The critical question then becomes: how do we optimize the specific behaviors that define organizational learning speed? To answer this, we must look beyond isolated tools and examine software delivery as a complex sociotechnical system.

Every complex system has points of leverage, and software delivery is no different. By taking a step back from individual tools and looking at the system as a whole, we can find the right leverage points to accelerate organizational learning.

The 2025 DORA report describes this need bluntly: “successful AI adoption is a systems problem, not a tools problem” (page 4). It also states, “when AI dramatically accelerates software development, our control systems—that’s us—must also speed up...We need fast feedback loops—faster than ever—to match AI-accelerated code generation” (page 9).

In this chapter, we’ll use systems thinking to illustrate the feedback loops that shape incentives, norms, workflows, and technical capabilities. We’ll look at what sociotechnical systems are, what levers shape them, and the essential role observability plays in creating effective feedback loops. You’ll learn which information flows to target—and why those produce better results than changing budgets, metrics, or org charts.

Sociotechnical Systems

Let’s start by using systems thinking to examine the software delivery ...