Why the people who can fix a company leave first

By month 12, Northstar’s VP of Engineering sees the company’s problems clearly: slow escalations, enterprise custom work becoming the norm, and top engineers quietly routing around issues instead of raising them. She treats it as a system problem, not a people problem, and changes what she can control.

Inside her organization, she starts rewarding early problem-spotting instead of punishing it. Engineers who flag ugly dependencies get protection and visible credit. Commitments are enforced more rigorously. The result is fast and measurable: within a quarter, escalation latency drops, bad news surfaces earlier, a key dependency problem finally gets handled, and morale improves.

The point of the essay is that the reform works — and still gets swallowed by the larger company. Northstar’s broader incentives stay the same. The enterprise pod still demands customizations, reporting upward is still cleaned up for the CEO, and the coordination layer still pushes escalations through a six-day path. At the boundaries with other teams, the old logic reasserts itself.

How a working reform gets absorbed

Her teams generate better signals, but those signals move into a system designed to flatten them into “fine.” The wider company never really sees that her experiment succeeded. What follows is not open sabotage but gradual erosion: each interface with the rest of the business weakens the standard she is trying to hold.

Recommended reading

SpaceX slips below IPO price after $1T wipeout

One small exception survives past month 15. Her practice of protecting the messenger of early bad news spreads into one adjacent org and persists there, apparently because one director independently believes in it and keeps enforcing it. The author is explicit that this is a person-dependent outcome, not a fully structural one.

That matters because the essay’s central claim is about selection. When an organization rewards adaptation over standards for long enough, it starts filtering who stays:

• people with the most options leave first
• then those with the strongest conviction
• then the informal, load-bearing people who uphold standards without formal recognition

What remains is not necessarily a bad team, but one shaped for high adaptation and low friction — calibrated to the gap between stated standards and enforced ones.

The warning sign leaders miss

The VP eventually starts taking recruiter calls, not because the experiment failed, but because she proved it worked and saw that the surrounding system made it unsustainable. That, the piece argues, is the real indictment: the people most able to correct the company are often the first to leave.

The practical signal to watch for is not obvious underperformance, but strong people burning unusual energy to produce ordinary-looking results. According to the author, that gap between effort and visible output is often the sign of someone holding a standard against structural resistance.

The exit interview may cite “growth” or “a new opportunity.” The deeper reality, the essay argues, is that the company had someone informally correcting its system — and left the system around them untouched until staying cost more than leaving.