Most of the time the decision gets made after the meeting. Nine people give up an hour and three of them say what they think. The real conversation happens that afternoon, in a thread with four people in it. The meeting produced a discussion; the decision happened somewhere else, later, with the wrong people present.
A recurring meeting keeps its slot long after it has stopped earning it, and once it is a habit nobody audits it. The first confident opinion sets the frame. Everyone who speaks afterwards is either agreeing with it or arguing against it, which is not the same as saying what they think. Disagreeing out loud costs something; disagreeing afterwards costs nothing, so the objection goes to the corridor. Asking the whole room to commit privately, at the same time, changes that price: the disagreement belongs to the meeting rather than to a person.
Six questions, most of a decision meeting. You would not run all six every week: one and six are worth keeping permanently, and two to five are for the week there is a real decision on the table. The stance question comes before the discussion, and that is the part that does the work.
A recurring meeting's real output is whether last week's decision survived the week. If a third of the room picks the wrong answer, that decision was discussed rather than made. Score the room, not the people in it.
Open text rather than a list, because a list of options tells people the answer. 'Decide' and 'update' appearing in the same cloud is the finding: half the room came to a different meeting.
Everyone commits at once, so the first person to talk cannot set a frame for the rest. Reservations get their own option deliberately — without it, people pick yes and raise the objection in three weeks.
The question means what it says. Anonymous, and asked while the decision is still open, is the only moment at which a corridor conversation is any use to the room.
A meeting that closes with four equally urgent actions has closed with none. An ordered list survives contact with next week; a flat one gets sorted by whoever is least busy.
One tap, and hard to argue with afterwards. A room that has just agreed something and rates its own follow-through at two has told you the decision is not actually made.
Asked at the end, it measures who won the argument. Asked at the start, it shows you the room you are actually in — including the three people who were never going to say so.
Add the stance question twice — once before the discussion and once at the close. Run the first copy, have the argument, then run the second. If nothing moved, the discussion changed nobody's mind and the meeting can be shorter. If it moved, that is the meeting earning its hour. Both rounds are saved, and running the second copy closes and freezes the first automatically, so you can read them back one after the other in the summary once the session has ended.
Six in favour and four against is not broad agreement. In a decision meeting the number that matters is how many people are on the other side, and an average is designed to hide exactly that.
Type the decision as you intend to write it up, then ask whether that is what was decided. The two minutes it costs are cheaper than the thread that follows a decision nobody can quote.
The one-word question takes half a minute. A room that cannot say what it is deciding is having a status update, and a status update does not need all nine people or all sixty minutes.
Every question above, ready to edit. Put your own decision in, drop the join code on the slide, and let the room answer from their phones.
Retro questions the quiet half of the team will actually answer
Open a meeting, workshop or conference session in ninety seconds
Brainstorming that collects ideas from the whole room, not the three people who like talking
Let the question everyone is thinking get asked without anyone having to own it
Training feedback that names the step where the room lost you
Measure what a course changed, not whether the room enjoyed it