Every organization I have spent time in has at least one recurring meeting whose origin nobody can trace.

It is usually a coordination meeting. Two teams, forty-five minutes, every second week. Ask why it exists and you get a slightly awkward answer about alignment. Ask who set it up and you get the name of somebody who left in 2022. Nobody defends it and nobody cancels it, and on the one occasion somebody did cancel it, two things went wrong the following month and it came back stronger than before.

I do want those meetings gone. What changed is what I think they are for. They grew over something, the way scar tissue does, and what they grew over is usually the same thing. Somewhere upstream, two teams ended up looking after the same thing on different time horizons — one building the future of it, one keeping the present of it running. Nobody decided this; a reorganization moved a line, or a platform team was created, or a capability got split down the middle of two managers who both wanted it. On an org chart it looks efficient, even sensible. In practice both teams are right about their half, neither is responsible for the whole, and every question that falls between them has nowhere to go.

So somebody, quite reasonably, sets up a meeting.

And it works. That is the part I used to walk straight past on my way to the diagnosis. Questions that had nowhere to go now have somewhere to go, once a fortnight. Two teams who would otherwise learn about each other’s plans afterwards learn about them in advance. Something real is being held together in that room, which is exactly why cancelling it broke two things the following month.

What the meeting is doing, underneath all of that, is clearing uncertainty. Nobody in the room can say with any confidence who decides what, so every second week they establish it again by hand, for the next fortnight’s worth of questions. That is a genuine purpose and it is genuinely being served.

It just costs those forty-five minutes permanently, from people who mostly cannot tell you why they are there. And because it works well enough, the uncertainty underneath it never becomes urgent enough to settle. The meeting takes exactly enough pressure off to keep itself necessary.

That is the pattern this chapter is about, and it generalizes further than meetings. When people cannot tell who owns an outcome, they do not stop working. They compensate. They build committees, approval chains, escalation paths, a shared spreadsheet with an owner column that everybody fills in differently. None of it comes from incompetence. It comes from uncertainty, and uncertainty is remarkably good at building things in its own defence.

The compensations are also, individually, correct. That is what makes them so durable. Each one solved a real problem on the day it was introduced, and removing it without addressing the thing underneath just reopens the injury.

Ownership is one of those words organizations use constantly and mean six different things by. Sometimes it means being the person left holding the outcome. Sometimes it means authority — someone who gets to say yes. The version this paper means is narrower and more useful than either:

Ownership is knowing, without asking, what’s yours to decide.

Which gives the shortest version of everything above, and the one line from all this that has followed me across twenty-five years:

Governance grows in exactly the space where ownership is unclear.

The organizations I have seen with the least of it were not disciplined about removing it. They had left it nothing to compensate for.

That kind of knowing doesn’t come from a title. It comes from a small set of agreed principles a team can lean on without asking — guardrails, which is where Part III of this paper picks the thread back up in full.

Purpose decided where the roots grow. Ownership decides who tends which one — and a root with no one tending it does not fail dramatically. It simply stops thickening, in a part of the tree nobody is looking at.

In every version of this I’ve watched happen, ownership created progress when it was taken, not when it was assigned. Nobody waited for a memo to feel ready — they took it, usually before anyone above them had noticed it needed taking, and the progress showed up afterwards, as the result rather than the cause.

Nobody ever scheduled the meeting that grew here. It grew because a question had nowhere else to go.

What moves this is not removing the meeting, which is the first thing everyone tries and which reliably brings it back within a quarter. It is settling the question underneath it — naming one owner for the thing two teams are each half-holding, specifically enough that a stranger could tell you who it is. Once that is settled the compensations stop being load-bearing and can come down without anything falling over. Done in the other order, you have removed the scar and left what caused it.

Cultivating this: Settling the Few Things, in Part III.