When the only thing shared is the hour

What it feels like from inside

• Everybody is busy and nobody knows what anybody else is doing, beyond a one-line summary heard weekly and immediately forgotten because it cannot be acted on.
• The meetings feel performative, and the reason is structural rather than cultural: the only thing a meeting can do for a group with no shared work is report, and reporting to people who cannot act is performance by definition.
• Asking a colleague for help feels like an imposition, because it is a cold request to somebody with no context and no stake.
• Everybody individually suspects the problem is them. Each person can see that their own relationships with colleagues are thin, cannot see anybody else's, and assumes the thinness is specific to them.
• And the team is described, by everybody including the manager, as a good team full of good people that somehow does not quite gel. That sentence is almost always accurate on both counts.

Why it is not a culture problem

The usual response is to treat it as one. The team does not feel like a team, culture is what makes a team feel like a team, therefore invest in culture: an offsite, a set of values, a weekly social, a ritual.

None of those put anything in the middle of the pyramid. They add rows at the top, where there are already two, and the top rows are exactly the ones that do not produce anything. A team with three wide administrative rows instead of two has had an expensive quarter.

The hole in the middle is a hole where shared artefacts, shared obligations and shared consumers would be. Those are properties of how the work is arranged, not of how people feel about each other, and they are changed by arranging work differently.

The two readings

There are two quite different situations that produce this shape, and telling them apart is the whole decision.

One: the work genuinely does not overlap. Eight specialists serving eight different consumers. Nothing is being paid for at the seams, because there are no seams. This group is correctly a list of people, the hole is not a defect, and the right response is to stop buying cohesion and spend the money on something else. There is a note on exactly this case in the first part of the site.

Two: the work does overlap, and the overlaps are being handled expensively — the same problem solved twice, decisions landing on people who did not know, outside parties getting inconsistent answers, handovers that go wrong. Here the hole is costing money and the fixes are small, specific and available.

The way to tell is the seam audit in the previous note. It takes an afternoon and it produces a list. An empty list means the first reading. A list of five means the second.

Filling the middle

One row at a time, aimed at a named seam.

The cheapest rows to add are obligations, because they recur without maintenance: a review roster, a duty rota, a shared on-call week. Second cheapest are artefacts that two people both change. Most expensive and most effective is a joint deadline.

What matters is that each addition is aimed at a pair that the seam audit named, rather than at the team in general. A row added at the team level lands at the top of the pyramid with the manager and the all-hands. A row added between two people lands in the middle, where the hole is.

Why it persists

Because every available response makes the top of the pyramid wider.

A team that does not feel like a team is a visible problem, and the responses within reach of most managers are all of one kind: add a ritual, add an event, add a shared statement of how we work. Each is cheap, each is immediate, and each lands at the top where there are already two rows.

The responses that would fill the middle — a rota, a shared artefact, two people given one deadline — require changing how work is assigned, which is slower, less visible, and produces nothing anybody can point at for a month.

So the pyramid gets wider at the top, year after year, and the hole stays exactly where it was.

The line that was refused

One of the rows given to the panel above claims nine people out of eight. It came from a list that still had somebody on it who had left two months earlier — the kind of error that is invisible in a spreadsheet and obvious the moment somebody tries to draw it.

It is worth saying because it is the ordinary case for this whole method. The value of drawing a thing rather than tabulating it is not that the picture is prettier. It is that a picture cannot be drawn at all when the numbers do not hold together, and a table can.