Every group has someone who barely speaks. Every group also, privately, has a theory about them: shy, disengaged, nothing to add. The theory is almost always wrong. Ask that person afterwards, one to one, what they thought of the discussion, and you will usually get the most considered answer you have heard all day.
So the question is not how to fix the quiet person. It is how to stop running meetings that only fluent, fast, confident speakers can get into.
Why doesn't calling on people work?
Because “Anything to add, Dana?” — dropped into a room where four people have already spoken fluently — is not an invitation. It is a spotlight, with thirty seconds to perform under it. The rational response is the one you always get: “No, I think that covers it.”
Being put on the spot punishes exactly the people it is meant to help, and it teaches them to prepare a deflection in advance. If your inclusion strategy is cold-calling, the quiet people are not included. They are braced.
What actually changes who speaks?
Separate thinking from talking. Most meetings fuse them: you compose your view in real time, out loud, while competing for the floor. That format favours speed and confidence, which are not the same thing as judgement.
The fix is five minutes of silent writing before any discussion. Everyone notes what they think first, privately. Then go around in a fixed order — everyone reads one item, no responses allowed until the round is complete. The floor is no longer something you have to seize; it comes to you, with your answer already in your hands.
Teams that try this are reliably startled by round two or three. The person who “never contributes” turns out to have been noticing things nobody else noticed, for months.
Why do narrow questions beat open ones?
“Any thoughts?” is the hardest question in the world to answer well. It asks the listener to survey everything, pick an angle, judge whether it is worth the room's time, and win the floor — all at once. Fluent talkers do this happily. Careful thinkers do the first three steps, decide the bar is too high, and stay quiet.
Narrow questions remove the survey step. “What was the most frustrating half hour of your week?” “Which part of this plan would you bet against?” “What did the customer actually say?” Each asks for a specific memory or a specific judgement, and specific requests get answers from almost everyone — because recalling is easier than editorialising.
Should you watch the airtime?
Yes, and not by feel. Everyone who runs meetings believes their meetings are balanced, in the same way everyone believes they are a good driver. Timed against a clock, most meetings are two voices and an audience.
You do not need software for this. Once a month, ask someone to tally rough speaking time on a piece of paper. The first tally is usually a shock, and the shock is useful — it moves the conversation from “we should include people more” to “Marco spoke for twenty-two of the forty minutes,” which is a fact a group can actually act on.
The counting works best when it is done by someone with nothing at stake in the result — which is exactly why the person who talks most should not be the one holding the stopwatch.
What if the quiet is actually about safety?
Sometimes the structure is fine and the silence is information. People do not share half-formed doubts in rooms where doubts have consequences — where the manager bristles, or the last person who disagreed is still being reminded of it.
You can test for this without a survey: run one written round unattributed. Collect the notes, shuffle them, read them aloud with no names. If the unattributed round surfaces things the named rounds never have, you have learned something important about the room, and no amount of facilitation technique will substitute for fixing it.
Common questions
Isn't it fine for some people to just listen?
Listening is fine; being structurally unable to enter is not. The test is whether they speak when the floor is handed to them in a fair round. Listeners pass with a short answer. Excluded people surprise you with a long one.
Won't fixed rounds make meetings slow and stilted?
For the first two sessions, yes — expect it and say so upfront. By the third it feels normal, and the meeting usually gets shorter, because you stop spending time on the same three people restating each other.
Does this work on video calls?
Better than in person, mostly. Writing works the same, rounds are easier to sequence with a visible list of names, and the chat gives quiet people a second channel. The one thing video makes worse is interruption recovery — so the no-responses-until-everyone-has-spoken rule matters more, not less.
What if one person still dominates despite the structure?
Then it is no longer a structure problem and it needs to be named — ideally by whoever is facilitating, as arithmetic rather than accusation: “You've had a lot of the floor today; I want to hear the rest of the room first.” It lands better from a neutral party than from a peer.
I'm building Facilitable, a facilitator for groups that meet regularly. It runs the written rounds, hands the floor around evenly, and notices who hasn't spoken — so nobody has to be the airtime police. It isn't ready yet. If you'd like to see it when it is: