Seeing Systems

Definition

Seeing systems is the habit of looking past the natural human urge to find a single, simple cause-and-effect chain for a problem and instead noticing that "every system is perfectly designed to get the results it gets" — that scattered, seemingly unrelated failures may be symptoms produced by one underlying structure. It pairs with an equally important discipline: distinguishing genuinely complex, interconnected problems (where root-cause thinking misleads) from merely complicated ones (where careful analysis of cause and effect still works).

In the Book

Yolanda and Doug apply this directly to the ten child-injury and -death incidents scattered across five bureaus and seven caseworkers that open the book. Doug's instinct is that there's "no pattern here" — just bad luck, an election year, and an economic downturn. Yolanda's counter-instinct, invoking the "perfectly designed" line, is that the incidents share a structural cause: something in how the system itself is built creates the openings for abuse and neglect, and it can be redesigned. The book develops the discipline needed to tell these situations apart in Chapter 2 through the Cynefin framework (David Snowden): in a complicated system you can research and find the correct cause-and-effect answer; in a complex system, cause and effect are only clear in retrospect, so the leadership move is not analysis but experimentation — probing the system to see how it responds, rather than diagnosing it once and prescribing a fix.

Why It Matters

Most organizational failures get responded to as either a target for blame (find who failed) or a target for analysis (find the broken step). Seeing systems adds a third read: the failure is an emergent property of how the pieces are connected, and no individual step is "wrong" in isolation. That reframing changes what interventions even make sense — it shifts effort from firing or fixing a person or a step toward changing the structure that keeps producing the same class of outcome, and it flags when a problem is too tangled for that kind of diagnosis at all and needs to be probed instead.