When something goes wrong in an organisation, the instinct is to find the part that failed and fix it.
A team missed a deadline, so the team gets a new process. A data pipeline produced wrong numbers, so the pipeline gets a new validation step. A stakeholder relationship broke down, so ownership gets reassigned.
Each of those responses is reasonable. None of them are wrong. But they share an assumption that I've come to question: that the problem was contained inside the thing that broke.
An organisation isn't a hierarchy. It isn't a structure chart or a set of departments with clearly defined edges.
It's a network of systems. And those systems are in constant relationship with each other in ways that don't appear anywhere in a diagram.
The relationship between a data team and a finance function is a system. It has its own history, its own norms, its own accumulated assumptions about who owes what to whom. Change something inside that system and it changes both parties, sometimes in ways neither expected.
The way a team is structured is a system. The meeting cadences, the communication habits, the informal channels where real decisions get made. These things shape behaviour just as reliably as any formal process does, often more so.
Most of the problems I've seen in organisations were visible at the level of one thing going wrong. The cause, when you looked for it, was usually somewhere else in the network entirely.
The part that broke isn't always
where the problem started.
This is why I've become wary of solutions that are too targeted.
A new tool, introduced into a team that doesn't trust each other, won't fix the trust problem. It will surface it faster. A restructure, imposed on a set of relationships that were already strained, doesn't reset those relationships. It relocates them, sometimes into harder-to-see places.
The system adapts. And it usually adapts in the direction of self-preservation, not in the direction the intervention intended.
That doesn't make change impossible. It makes the question before any change more important: what will shift elsewhere when we change this here, and are we ready for that?
The most effective leaders I've worked with or observed think this way almost instinctively.
They don't ask what's broken. They ask what's the system that produced this outcome, and what would have to be different for a different outcome to emerge?
It's a slower question. It resists the urgency to act quickly and visibly. But it tends to produce changes that actually hold, rather than changes that simply move the problem one step further from where anyone is looking.
Organisations are made of people. And people, like all living systems, are more complex than whatever map we drew of them.
Treating them that way is where good leadership usually starts.