Every company says they want innovation. Almost none of them mean it.

Most organizations are still running a 1911 factory management system on knowledge workers: reward visible activity, reduce variance, protect the chain of approval. If you’re treating knowledge work like factory work, you need workers who execute tasks without questioning them.

The person who schedules the meeting about the meeting gets rewarded. The person who questions whether we need the meeting at all gets labeled “difficult.”

Here’s what it looks like in practice. A team builds something genuinely new. Instead of protecting it, the organization routes it through the same approval chain that exists to de-risk everything else. By the time it survives stakeholder reviews and alignment meetings, it’s been sanded down into something safe enough to ship and boring enough to ignore.

Then someone in leadership asks why the innovation team isn’t producing results. The team responds with a deck. It has charts. Leadership schedules a follow-up. It happens next quarter.

But there’s something deeper going on. Most knowledge workers’ self-concept is tied to their title. Climb high enough and the title is the self-concept — ego is the north star. When someone on their team approaches things from first principles — questioning assumptions rather than inheriting them — that’s not an opportunity. That’s an existential threat. The note-taker gets promoted because they reinforce the leader’s importance. The innovator gets sidelined because they expose their inadequacy.

The incentive structure punishes the exact behavior the strategy deck celebrates. And incentives, not intentions, drive behavior. We don’t follow the mission statement. We follow the incentives. We don’t do what’s celebrated in the all-hands meeting. We do what’s safe on Tuesday morning. Most people never question whether there’s another way to keep score, because the current one is all they’ve ever been handed. The ones who have built their own scorecard are either the most valuable person in the room or the most threatening one, depending on who’s leading.

The real innovation I’ve seen almost never comes with approval. It comes from someone curious enough to build without asking permission. Usually in the margins. Usually without a roadmap. Usually at some personal cost — by someone who’d decided they had nothing to lose.

The leaders who break this pattern share one thing: they care more about the outcome than their position in the room. They assign problems, not tasks. They protect useful dissent and tolerate the friction that comes with genuine innovation. They’re comfortable letting other people shine, and secure enough not to confuse visibility with value. This matters more now, not less — AI is rapidly commoditizing task execution, and it doesn’t care about your political capital. The old Taylorist model is running out of road. Simple formula. All too rare. And because it’s harder to identify than the loud, performative kind of ‘leadership,’ it usually goes unnoticed.

I used to think organizations were broken. Now I understand many of them are doing exactly what they were designed to do.

The harder question is what you do once you see it.