Team building that changes something

There are two kinds of team building and they look identical in the photographs.

One fills an afternoon. Everyone is pleasant, the activity is competently run, people go home slightly early, and by the following Monday nothing at all is different. The other produces something a team is still referring to months later, usually a single moment where somebody did something surprising and everyone saw it.

The difference is not budget, and it is rarely the activity itself. It is whether the day was designed to make something visible.

The activity is the setup, not the product

A session changes how people work together only if two things are true.

The activity forces a group to decide something together under a real constraint. Time, materials, information, or physical coordination. Constraint is what produces behaviour. Without it, a group cooperates comfortably, learns nothing about itself, and the day is a pleasant outing that could have been lunch.

Somebody talks about how the decision got made. Not whether the team won. How they organised, who ended up leading, what got ignored, what the group did when it fell behind. This is the debrief and it is the entire product. An activity without one is entertainment, which is a perfectly good thing to buy, but it should be bought under its own name.

Ask a supplier what the debrief looks like. If the answer is a slide with the winning team's photo on it, you are buying entertainment.

Group size decides everything else

The single most useful number in this whole subject is the size of a working group. Five to seven.

Below five there are not enough different views for disagreement to be productive, so the group agrees quickly and nothing surfaces. Above seven or eight, the quieter people stop contributing. The group splits into two or three talkers and an audience, which is exactly the dynamic most companies are running the session to fix.

Everything else follows from that number: how many facilitators you need, how many stations, whether the format can be one activity or has to be a rotation, and whether the day's real risk is the activity or the queueing.

Three ways a session quietly fails

It is really a performance. Anything where one person does the task while the rest watch and cheer. Cooking classes often drift here: two people cook, six take photographs. Fun, but nothing is being built between people who watch each other.

One strong player carries it. If the activity rewards an existing skill somebody already has, that person wins it for the group and everyone else learns that they were right to leave it to them. Good activities are ones where nobody in the room has an advantage, which is why unfamiliar physical tasks and unfamiliar making tasks both work well.

The teams are the existing teams. Putting the finance team together as the finance team produces the same conversation they have every day. If the goal is to connect people across the company, the groups have to cut across it, and somebody has to actually assign them rather than letting the room self-select. People sit with the people they already know.

If the activity has no constraint, there is nothing to observe. If there is no debrief, nothing gets carried back.

Physical, cognitive, or making

Most formats fall into three families, and they suit different problems.

• Physical puts everyone at the same disadvantage and produces coordination fast. Best for groups that do not know each other yet. It also excludes people, so there has to be a real role for anyone not doing the physical part, decided in advance rather than improvised on the day.
• Cognitive puzzles, cases, simulations. These surface how a group handles information and disagreement, which is usually the actual problem. They need a facilitator who can read a room.
• Making building or cooking something for a purpose. Slower, quieter, and the strongest of the three for teams that are already familiar but not working well, because it takes several hours during which people talk without being asked to.

How long it needs to be

Half a day is the honest minimum. One activity, run properly, with a real conversation afterwards, takes about three hours once you count arrival and settling. Anything shorter and the first thing cut is the debrief, because it is the part that has no fixed end time and feels the most optional in the moment. It is not optional. It is the whole reason the day was booked.

A full day earns its keep in two situations. The first is when there are two genuinely contrasting activities, so people see each other in two different modes rather than the same mode twice. The second is when getting to the venue eats the morning anyway, in which case a half day was never really a half day.

What a full day should not be is the same afternoon stretched. Groups notice, and the second half of a padded day is where the phones come out.

One more piece of timing that gets decided by accident: where the session sits relative to lunch. A session that runs through lunch keeps its energy and its groups together, and people talk over the food, which is often where the useful conversations happen. A session split by a lunch break loses twenty minutes to reassembly and comes back as a different room.

What to give a supplier

The brief that gets a useful proposal is short. Headcount and the spread of ages and physical ability. Whether the group knows each other. One sentence on what should be different afterwards. Whether it has to be indoors. The budget.

That last sentence is the one most briefs leave out, and it is the one that decides everything. We ask it first: what has to be true when people are back at their desks on Monday. A supplier who is given that will propose something specific. A supplier who is given only a headcount and a date will propose whatever they run most often.