Every reorg begins with the same sentence: *we have a people problem.*
Sometimes it arrives as a performance plan. Sometimes as a culture conversation. Sometimes as a new org chart, drawn late at night, with certain boxes quietly missing. The diagnosis underneath is always the same — the people aren’t performing — and I have spent my career watching that diagnosis be wrong in the same direction, over and over.
Most performance problems are system problems wearing people-problem costumes.
—
Here’s why the costume works so well.
Symptoms are always local. A missed target shows up on *someone’s* dashboard. A dropped handoff dies in *someone’s* inbox. Low morale presents in *specific* people, in specific meetings, with specific faces. Pain appears in humans — it has nowhere else to appear — and so the diagnosis attaches to humans, with their names conveniently right there on the org chart.
So the person gets managed. Coached, planned, culture-conversationed, and eventually managed out.
And the system that produced the symptoms? Untouched. Fully intact. Waiting.
Which is why the next person in the same role produces the same symptoms — a little faster, usually, because the role’s unofficial second job got heavier during the transition. You met this pattern in Chapter 3 when Susan left, and in the last section as the backfill tell. Here it is as a law: **if multiple capable people, in sequence, struggle with the same thing, the system is the common variable.**
Three different humans do not fail identically by coincidence. Coincidence doesn’t repeat with that kind of discipline.
—
So the question changes. The one I bring into every engagement:
Not *”why is this person underperforming?”*
But *”what does it say about the system that capable people keep not doing this thing?”*
The first question ends in a termination and a repeat. The second one ends somewhere useful — because it finally puts the structure on the examination table instead of the human.
Why don’t we ask it naturally? Because blaming people is cheaper. Immediately cheaper, anyway. A performance plan costs a form. A systems examination costs an admission — that the structure leadership built, or tolerated, or inherited and never questioned, is producing these outcomes on purpose. The org chart doesn’t have a box for *the way we work*, so *the way we work* never gets put on a plan. People do.
Blaming the team is cheaper than examining the system. It is also why the same problems keep coming back.
—
Before the next performance plan, run this list. It takes an afternoon and it’s cheaper than a mis-fired employee:
Does this role have unambiguous asks, or is it paying the ambiguity tax we covered in Chapter 3? Do its handoffs exist by design, or by the goodwill of whoever’s on each end? Is it secretly two jobs — one written down, one absorbed? Was the predecessor also, curiously, “not a fit”? And if your honest answers make you wince — good. Wincing is diagnostic.
—
This closes Part Two, so let me say what Part Two was for.
The humans in your building are not malfunctioning. The ones with conditions are managing them while delivering. The founder’s grip is a memory doing its job. The heroes are absorbing what the structure drops. The “underperformers” are, more often than anyone wants to admit, capable people reporting a system failure with their symptoms — the only reporting channel the system left them.
The problem was never the people. It’s what you built them inside.
So the obvious next question: what actually *is* happening inside there — in the structure nobody’s been examining while everyone was busy examining the humans?
That’s Part Three. Bring a flashlight.