The meeting ended. The decision didn't.

Dispatch · to the desk of one working leader

The meeting ended. The decision didn't.

The observation

Every decision is made twice.

Once in the room, where everyone nods and the notes look clean.

And once by the elevator, where two of the people who nodded agree to wait and see.

You only run one of them.

The pattern

Yes in the room. Nothing moved by quarter end.

Here is why.

A decision has two parts. The official part: agenda, notes, action items. The human part: the private moment after, when people decide if they meant that yes.

We keep the first part and cut the second. Nobody is being lazy about it. The calendar is booked wall to wall, and the short talk that would have made the yes real is the first thing to go.

That talk is where the real questions live. Who tells the night shift? Can my team do this by March? Questions like those rarely make it into the notes. They get asked by the elevator instead, quietly, to one person.

The manager who knows March is impossible says nothing, because the room rewards speed, not doubt. Six weeks later the slip arrives, and it surprises everyone but her.

Read that as sabotage and you will manage the wrong thing. The elevator is simply where truth goes when the room will not pay for it. You do not have a meeting problem. You have a truth-timing problem.

Erin Meyer, a professor at INSEAD, maps this on what she calls the Deciding scale in The Culture Map. In consensual cultures, the decision is built before the meeting, through quiet agreement. In top-down cultures, the boss decides, and the decision stays open to change.

Japanese companies have a name for that groundwork. Nemawashi: the quiet one-on-one work done before the meeting, so that by the time everyone sits down, the decision is already real. The meeting just confirms it.

The word comes from gardening. It means going around the roots. Before you move a tree, you prepare its roots for weeks, so the move does not kill it. A decision travels the same way. Prepare it in private, and it survives the room.

We do it backwards. We decide in the room, then unmake it by the elevator.

Now the new part. AI writes the agenda, the notes, the action items. The official half of deciding is becoming free, and it is getting very good. What it cannot do is stand in the corridor and notice who went quiet.

And the better the machine gets, the worse the trap. A flawless plan reads like a real yes. It is only a real document.

Hold this line

Guard the ten minutes after the hour. That is when people decide whether they meant it.

The move

You cannot rebuild nemawashi by Friday. But you can stop cutting the ten minutes where it would have lived.

Once this week, do not end your most important decision on the hour.

Stay ten minutes. Screens off. One question: what would make this hard to deliver? Then listen.

If nobody speaks, ask once more and wait. Silence on the first ask is politeness, not agreement.

And when the concern lands, do not fix it on the spot. Write it down where everyone can see it. A concern you record turns into a plan. One you talk over just moves to the corridor.

Book the room longer up front, because intentions lose to calendars. Walk people out. Let the call breathe.

The week after, put that concern back on the table once. Ask what has changed. That follow-up is the whole thing. Skip it and your team files the ten minutes under theater.

If ten minutes feels impossible, start with five. What matters is that the yes gets some air before people leave the room.

You did not shorten the meeting.
You shortened the agreement.

Which yes from this quarter never made it past the elevator? One sentence. I read every reply.

See you Tuesday.

Rui

P.S.  Worth your margin: Erin Meyer's The Culture Map, the chapter on deciding. One flight, whole chapter.