Judgement Comes from Density

After the last piece on the elders, a friend pressed us: with so few people, what happens when one judgement is wrong?

Our position first. It is precisely because there are few of us that a judgement is unlikely to be wrong. Why? Because judgement comes from the context density of the organisation, not from one person's lone nerve. Context density means the same real thing sitting in several heads at once. On this one we are fairly confident.

People are the variable, context is the constant

Organisations usually grow by adding people. We are trying the other route: hold the headcount still and put what grows back into the context. Thirty-six colleagues, six theatres, six SEAL teams, six people to a team on average — one meeting room holds them, and one meeting can finish the conversation. So why insist on holding it still? Because past one meeting room, candour starts taking a discount. Some things can be said clearly in private and not in public: you say five, and the listener fills it in to fifty. The other forty-five is guesswork.

Draw a curve. Headcount on the horizontal axis, the real transactions each person touches in a year on the vertical. It rises, then falls. We are standing at the turning point. Go further right and there are more people, but what each of them holds is less and less real. Twenty to thirty mandates a year, half new and half repeat mandates, spread across thirty-six people, and everyone is still next to the table. Headcount is the denominator, real transactions are the numerator. Double it? Half the firm is left with reporting.

Every project is best carried by one person from end to end

Understanding of a project is continuous; change the person and it breaks. So there is no assembly line inside a SEAL team: the person who owns it is there at D01 when the fact base is laid, and there at D60 when the money settles. Sixty days, eight milestones, not one missed. The investment committee can argue, but the owner has to sit through all of it, or the judgement cannot be assembled. Clients say we are diligent and can see through the froth of an industry. That is not really about attitude. It is that we never swapped the person out.

There is a rough line we use inside: everyone is a carbon-based agent, and an agent has to be fed context too. A pit one project falls into tonight should be another project's solution by tomorrow morning. First, it saves someone else the time of falling in again; second, it keeps eight years of experience inside the organisation rather than on one person's CV. If experience only settles on individuals, the organisation has no compounding. Where density is thin, compounding breaks.

So can we add people at all? We can. Our people can grow in number, our mandates cannot. Adding people adds capacity to stand beside companies; adding mandates dilutes a promise already given. Limited capacity is not a pose, it is arithmetic: judgement, research, coordination and a partner's attention are all non-renewable. Spread across a hundred projects, it means breaking your word to a hundred projects.

The real ceiling on a firm's size is not how many people it can hire, but how many people it can keep inside the same context. Taking on few is so that everyone is still standing at the table, not so that we look restrained. Where exactly that ceiling sits is not answered on paper; it is answered in the mandates, year after year. As far as we can tell, we are standing right beside it.