The Peter Principle

Early in my career, I worked with a senior developer who was genuinely exceptional at his craft. He solved hard technical problems that made other engineers give up. He was promoted to engineering manager. Eighteen months later, he was gone — burned out. The work he'd been exceptional at had disappeared from his days. The work that replaced it, he'd never asked for.

That's the Peter Principle in practice, and it's more brutal than it sounds in theory.

Laurence Peter's 1969 observation was this: in hierarchies, people get promoted based on how well they do their current job, not their next one. So they keep rising until they land somewhere they can't perform. Then they stay there. The organization settles into a kind of managed dysfunction — senior positions filled by people who peaked one job ago.

But here's what doesn't get said enough: the problem isn't the people. It's the reward structure.

We built organizations where the only way to earn more money, more respect, more status is to stop doing the thing you're good at. An exceptional engineer who doesn't want to manage anyone has two options: get promoted into management, or watch less talented colleagues earn more and advance further. That's not a career ladder. That's a trap with a bow on it.

The competence assumption is the other piece — the silent belief that domain expertise transfers automatically to leadership. It doesn't. The skills that make a great engineer are often the inverse of what makes a great manager. Precise thinking, tight control over variables, preference for clear-cut answers — those are engineering virtues. Management is messier. You're herding ambiguity, not eliminating it. You're reading people, not systems. Some of the best engineers I've worked with became awful managers precisely because they couldn't tolerate the uncertainty that management requires.

So we end up with a double loss. The star engineer disappears into meetings they hate. The team they now lead gets someone who was never trained to lead, resents the job, and has lost the technical authority that made them credible in the first place. Everyone loses. And then we wonder why retention is hard.

The fix most organizations try — give managers leadership training after they're already managing — is backwards. By the time you're in the role, the damage is already spreading.

What actually works is boring but rare: parallel tracks. Let engineers advance in title, compensation, and status without ever managing a single person. A Principal Engineer should earn as much as a Director. Some people are genuinely wired for people leadership — find them early, look for evidence rather than seniority, and invest in them before the job starts. Look for who's already mentoring, already smoothing conflicts between teams, already thinking about priorities at a level above their own ticket queue.

Build in a real escape hatch, too. The manager who realizes after six months that this isn't for them should be able to step back to individual contributor work without shame. Right now, stepping back feels like failure. It gets treated as a demotion. It shouldn't. It's just information — information that came a little late, but is still worth acting on.

I've made the mistake of promoting for loyalty and tenure rather than fit. I've also watched engineers rediscover their hunger once they were no longer trapped in a role that didn't suit them. The change in output was immediate and visible — not because the person had changed, but because the mismatch had finally been corrected.

The Peter Principle isn't really about people rising to incompetence. It's about organizations that offer only one path up, and then act surprised when people take it even when it leads somewhere they shouldn't be.