Forty People Stood in a Room and Nobody Was Allowed to Speak

Someone from another department walks up to you in your second week and says, casually, that any of the teams can pick up this piece of work — except one. Not that team. Anything but that team. And you realise you’ve just been handed a piece of information nobody wrote down anywhere: one part of your organisation has quietly lost the trust of the people it serves. What do you do with that on day nine?

Yves Hanoulle — who describes himself as a “creative collaboration agent” rather than an agile coach — told Andrew Harmel-Law, Andrea Magnorsky and Kenny Schwegler about a day roughly twenty years ago when he responded to that moment by stopping all work for forty people and moving the furniture out of the biggest room in the building. The interesting part isn’t the outcome. It’s what happened in the two hours before people were willing to actually decide anything.

The complaint that contained the answer

Yves brought everyone together and ran what he calls “some kind of retro thing.” The team quickly agreed the four-team structure was wrong, and that three teams would be better. Around 60% of the small groups landed on the same conclusion independently. Easy part done.

But the diagnosis came with something else attached. People said the original four teams had been badly formed — decided by Yves’s predecessor together with the department head. Lots of finger-pointing at his boss, who wasn’t in the room, and at his predecessor, who had left and couldn’t defend himself.

Yves noted it and moved on. He didn’t argue. He didn’t defend anyone. He just filed it away, and it became the most useful thing he heard all morning.

Chairs out, tables out, mouths closed

Over lunch, Yves booked the largest room he could find for the whole afternoon and, with his boss, cleared out every chair and table. In a large corporation, moving furniture is itself a minor political act — his boss handled the shielding, on a floor neither of them had really seen before.

Then he told them the rules. “I told you when I came here that I would help you make decisions, but I would not make decisions for you.” Walk around in silence. Gather yourselves into three circles. When you’re happy, stop moving.

So thirty to forty people shuffled around a bare room, not speaking. A little movement, then stillness. Yves would ask: are we happy? No. Move again. Between rounds, talking was allowed — questions, short statements — then back to silence. “Because they were not allowed to speak, it really was feeling the things.”

He also refused something they badly wanted. People asked for a list of requirements up front: does every team need an architect, a tester, a senior? Yves said no, and they hated him for it. His reasoning was practical — agreeing that list would take days. “You look for yourself. What is important for you?” He and Kenny might disagree on the details and still end up in groups they’re both happy with. Why force an agreement nobody needs?

“Yves, this is not working. You should decide.”

Twenty minutes in, someone pointed at him and said exactly that. And it was, of all people, the loudest voice from the morning — the one who had complained hardest about the boss making this decision badly the first time.

Yves looked him in the face and said no. He’d been there two weeks. He barely knew people’s names. And he pointed straight back at the morning’s complaint. “I could just see clicking and looking in his eyes that it clicked, and I saw some other people nodding.”

After that, the room moved much more. Andrew’s read on it later: people say “I hate this decision,” but when you hand them the pen, what they actually meant was “I want someone to blame.”

What the room showed that nobody had said

Yves’s boss, watching from the side, spotted something. The people from the problem team — the yellow one nobody wanted work from — had all clustered into a single group and weren’t moving.

There was a reason. This is Belgium. Around four people in the room were French-speaking; the rest spoke Dutch. Of course they grouped together. One of them was more senior and seemed to be mentoring the other three.

When another strong voice complained that “they are all sticking together, and they don’t want to change,” Yves noticed the person was talking to him, not to the group. He answered directly: I can see that, and everybody can take responsibility — if you don’t like it, you can move in there. It clicked again. That person moved, the cluster loosened, and people started rearranging.

Then the sharpest moment of the day. One person left the French-speaking group and tried to join another circle. Two people there physically closed ranks and blocked them out. Nothing was said, because nothing could be said. “I’ve never seen something so visually clear like, ‘I don’t wanna work with you.’”

Yves didn’t name it in the room. He wasn’t sure how to. The blocked person moved elsewhere and the situation resolved itself, in a manner of speaking.

The person who cried, and the seniors who split up

At some point one person was in no group at all and started crying — the decision felt too hard. They stepped outside with Yves’s boss, Geert, and came back about five minutes later. Yves wasn’t in that conversation, but from the debrief, Geert’s message was roughly: you want to join that group but you’re not sure they want you — put yourself there and see.

This was right after everyone had watched someone get blocked. So the fear was reasonable. But when this person walked into a circle, people made space. Yves saw the smile arrive. “Really feeling like, ‘Oh, I’m welcome here.’”

And in the last round, something Kenny had been probing for actually happened. The two most senior people, who had stayed together through most of the exercise, split up and went into different teams. Not because anyone told them to. Because one of them worked out the whole thing would be better with a senior in each group.

They finished at four o’clock. Forty people, no software written all day, and Yves then spent the following days defending that with his boss — most loudly to the internal customer who had started the whole thing. His own assessment: by the end of that sprint they delivered more than they would have if they’d kept working. The team composition then held for over a year, including through a round of layoffs.

What made this work, and what it cost

• Deciding that a decision is needed is itself a decision. Andrew’s point in the episode: Yves didn’t decide the teams, and he didn’t decide that he or the boss would decide. He decided that the group would, and he decided how. That’s a narrow but real form of authority, and using it well is what made the rest possible.

• The complaint became the constraint. Because people had spent the morning saying “managers shouldn’t decide this,” Yves had something concrete to hold up when they asked him to decide it. He didn’t need to argue from principle. He argued from their own words, an hour old.

• Removing speech surfaced things speech was hiding. The language split, the mentoring relationship, the personal conflict — none of that would have come out in a discussion. It came out as bodies in space. That’s also the risk: a physical exercise makes rejection physical too.

• Refusing to define the criteria up front was the unpopular call, not the popular one. Everyone wanted a checklist. A checklist would have taken days and produced an agreement nobody actually needed. People can optimise for what matters to them individually and still land somewhere collectively acceptable.

• Two facilitators, not one. Geert was the boss, which cuts both ways — people may feel less safe. But he could watch patterns Yves missed, take the crying person outside, phone the two people on holiday, and shield the furniture-moving politically. Yves was clear he wouldn’t have wanted to run this alone.

• The absent people were a genuine gap. Two people were on holiday, represented by chairs. Yves offered them a chance to change things afterwards and was quietly relieved when they declined, because he had no good answer for how to redo it.

• Facilitation of this kind is exhausting even when you barely speak. “I haven’t — didn’t have to do much, but I had to do a lot.” Worth budgeting for.

The part nobody had a plan for

The blocking incident is the thing that sticks. Andrea pushed on it near the end: that’s a very strong social signal, so what did you do with it afterwards? And the honest answer was not much — Yves was busy defending the lost day, and it took him twenty years and a chance conversation to find out that the blocked person is still at that company, still someone who “needs a manual to work with,” though considerably easier now.

Which is what makes the story useful rather than tidy. A well-facilitated process can make invisible dynamics visible in about ninety seconds. It doesn’t tell you what to do with what you’ve seen. If you run something that gets the truth into the open, the question worth asking before you start is: what’s my plan for the thing I wasn’t hoping to find out?

Further Reading

Mentioned in this episode

• Conway’s Law — raised by Yves as the reason team composition isn’t a purely social question: how you organise people shapes the software they produce.

• The Peter Principle (Laurence J. Peter) — Andrew’s shorthand for people promoted into management on the strength of technical work, then handed responsibilities like pastoral care that nobody trained them for.

Worth exploring

• Team Topologies by Matthew Skelton and Manuel Pais — the modern treatment of team boundaries as a design decision with architectural consequences, which is exactly what that afternoon in the empty room was.

• Dynamic Reteaming by Heidi Helfand — a whole book on how and when to rearrange teams, including self-selection approaches very close to what Yves ran.

• Creating Great Teams: How Self-Selection Lets People Excel by Sandy Mamoli and David Whitehouse — practical mechanics for running self-selection at scale, useful if you want the version with more scaffolding than “walk around in silence.”

• The Skilled Facilitator by Roger Schwarz — on naming what you observe in a group, including the uncomfortable things, which is precisely the gap Yves admits to in the blocking moment.