# Every org chart is a theory. The work is the experiment.
An org chart is a hypothesis about how work flows through an organization. It's a guess — sometimes an educated one, sometimes not — about who needs to talk to whom, who decides what, and where accountability lives. Like all hypotheses, it needs to be tested against reality. And like most hypotheses, it turns out to be partially wrong.
The problem is that most organizations treat their org chart like a fact instead of a theory. They design reporting structures in a conference room, draw the boxes, send the all-hands email, and then expect the work to flow along the lines they drew. When it doesn't — and it never entirely does — they blame the people instead of the map.
The org chart is a map. The work is the territory. When they disagree, trust the territory.
The shadow organization
Every company has two organizational structures. There's the official one — the boxes and lines, the reporting relationships, the titles. And there's the real one — the informal network of who actually calls whom when something needs to happen.
I worked with a manufacturing company where the org chart said quality issues went from the floor supervisor to the quality manager to the plant director. The actual path was: floor supervisor texts Maria in engineering, Maria walks down to the floor, they figure it out together, and then Maria emails the quality manager a summary so the official record looks right. The org chart described a three-step escalation. The work described a two-person collaboration with a reporting step tacked on for compliance.
Maria wasn't circumventing the system. She was the system. The org chart just hadn't caught up.
This shadow structure exists in every organization I've ever worked with. People find the shortest path to getting things done, and that path almost never matches the official flowchart. Sometimes the shortcuts are brilliant — they represent the organization's accumulated intelligence about how work actually moves. Sometimes they're dangerous — they create single points of failure or bypass important checks. Usually they're both.
Why the theory is always wrong
Org charts fail as maps for a few predictable reasons.
First, they assume work is hierarchical. Most interesting work isn't. A customer problem might touch sales, engineering, logistics, and finance. The org chart says those are four separate branches. The work says they're one team for the next forty-eight hours. The org chart doesn't have a way to represent temporary, problem-specific collaboration. So people improvise it, and the improvisation looks like chaos to anyone reading the chart.
Second, org charts optimize for clarity of authority, not for speed of information. The chain of command might be perfectly clear, but if a decision requires approval from three branches, the information has to travel up one tree, across the canopy, and down another tree before anything happens. In the time that takes, the person closest to the problem has already solved it using their phone and a favor.
Third, org charts are static. They're designed for steady-state operations. But organizations spend a surprising amount of their time in non-steady states — launching something, recovering from something, responding to something. The org chart describes the organism at rest. The work reveals the organism under load. Those are different animals.
Reading the work
If you want to understand how an organization actually functions, don't read the org chart. Read the work.
Watch where emails cluster. Notice who gets pulled into meetings they're not "supposed" to be in. Pay attention to who people call when something breaks at 4 PM on a Friday — that's the real leadership structure, stripped of politics and ceremony.
Look at handoffs. Every time work passes from one person or team to another, there's a translation moment. Information gets summarized, reformatted, sometimes lost. The number of handoffs in a process tells you something about the org chart's accuracy. More handoffs than the work requires usually means the structure is creating steps the work doesn't need.
Look at where decisions stall. If a decision sits in someone's inbox for three days, that's not laziness — it's the org chart assigning authority to someone who doesn't have the context to decide quickly. The work is telling you that the decision-maker is in the wrong box.
Redesigning with humility
I'm not against org charts. Every organization needs structure. But the structure should be a response to how work actually flows, not a prescription for how it should flow.
The best organizational redesigns I've been part of started by observing the shadow organization and then formalizing the parts that worked. Instead of drawing new boxes in a conference room, we watched where the work naturally clustered, who the real connectors were, and which handoffs added value versus which ones just added delay.
Then we drew the chart to match reality, not the other way around.
It's less dramatic than a big reorg announcement. It doesn't generate the kind of slides that look impressive at a board meeting. But it works, because it starts from what's true instead of what's theoretical.
Every org chart is a theory. Treat it like one. Test it against the work. Revise it when the evidence says you're wrong. That's not organizational instability — that's organizational honesty.