Part III — Cultivating Organizational Flow · Chapter 38 · 6 min read · First Public Draft

Human conditions

Growing What Cannot Be Installed

Trust, empathy and courage do not respond to programmes. They respond to what happens in specific small moments, repeatedly.

Reduces frictionCreates valueOwnership clarityHandoff lossDecision latencyLearning cycle time

Nobody Writes Down the Cost of Not Trusting argues that these three conditions are load-bearing, and that their absence hardens into structure. This is the chapter about growing them, which is slower than anything else in Part III and has the worst reporting.

The first thing to say is what does not work, because it is the thing that gets tried. A values programme does not produce trust. An offsite does not produce it, though people often come back feeling as if it did, which is its own problem — the feeling fades in a fortnight and the belief that something was fixed lasts a quarter. A statement of behaviours on a wall produces nothing at all except a small, permanent tax on everyone’s credulity.

None of that is because the people running them are naive. It is because these three conditions are not beliefs to be installed. They are conclusions people draw from evidence, and the evidence is what actually happened the last few times it mattered.

Trust is built in the small transactions, and it has a bad payoff schedule

Trust grows the way Ten Minutes Today describes: people repeatedly helping each other move, until it stops being a pleasant surprise and becomes a reasonable expectation. Slow accumulation, fast loss, in that order.

What is worth adding here is where the leverage sits, because it is not spread evenly.

Remove one check and let the result stand. The strongest signal available is somebody senior taking away a control and then not reinstating it the first time something goes slightly wrong. That single act says more than any amount of stated intent, and it is expensive precisely because it is a real risk taken in public.

Let people meet the consequence of their own work. Most distrust between teams is distrust in the abstract — a category, not a person. It survives on distance. It rarely survives two people who have had to solve something together at a whiteboard.

Watch what happens to the person who admitted the mistake. That is the whole test, and everybody in the organization is quietly running it. If the answer is a process, you have just bought a control and sold the honesty that would have caught the next one earlier.

Empathy is mostly a matter of who has seen whose week

Empathy is a feeling, and I am not going to define that away — it is a genuine willingness to care what somebody else’s week is like, and some people simply have more of it than others.

What an organization can do something about is the other half. Willingness does not get you far without a picture, and most of the time the willingness is there and the picture is missing: people cannot see the constraints of a team they have never sat with. They are not indifferent. They are guessing, and guessing badly, about a week they have never spent.

Which makes it one of the few things on this list with a mechanical fix. Guest of Reality, in Creating Value, is the version I have seen work — put people where the work lands, for a day, on purpose. It fixes small things immediately, and it settles the estimating in both directions, but the durable effect is that the other side stops being a category.

The same move works between internal teams and is almost never used there. Somebody from the team that files the requests spends a morning with the team that receives them. Nothing is presented. They just watch the queue arrive.

And the smaller daily version: when a team says we are at capacity, treat it as information about their week rather than as a negotiating position. That one sentence, taken at face value a few times, does more for a boundary than any service-level agreement.

Courage needs the cost taken off the person showing it

Courage is the one where the organization has all the leverage and rarely uses it, because it is entirely determined by what happens in the ten seconds after somebody says the awkward thing.

Go first, if you are the most senior person in the room. Everyone Left Agreeing is a problem of who speaks first, not of who has doubts. The doubts are already there. Somebody has to make the first one safe, and the cost of going first falls almost entirely on whoever has the least to lose — which is the person who least needs the protection and therefore the person who should do it.

Thank the objection separately from deciding about it. These get conflated, and the conflation teaches people that raising something only counts if it wins. Say the objection was worth having, then decide against it if that is the decision. Both halves matter.

Look at what happened to the last person who was right too early. Every organization has one. Whether they were promoted, ignored, or quietly moved is the most accurate statement of policy the place has, and nobody wrote it down.

Keep somebody who is willing to go first, and make it survivable

Nobody Writes Down the Cost of Not Trusting ends on the organization that had a trail-breaker, found them uncomfortable, and taught them to stop. That is the failure worth designing against, and it is a staffing question as much as a cultural one.

Notice who actually goes first, and check they have not gone quiet. Every organization has one or two. They are usually the people who cost the most attention per week and who somebody has, at some point, described as difficult. Losing them is silent, because what disappears is proposals that were never made.

Pair going-first with bringing-along rather than treating them as two kinds of person. A trail-breaker with no empathy for the pace behind them produces resentment and a route nobody follows. Somebody who only consolidates produces a very tidy version of where the organization already was. The useful move is putting the two together on the same problem often enough that each stops finding the other’s instinct unreasonable.

And protect the retreat. Breaking trail means being wrong about the route in public. If a wrong route costs somebody standing, you will get one attempt and then no more — not because they lost their nerve, but because they read the evidence correctly.

The part that makes this hard to run

None of these produce anything you can report. There is no number that moves, no artefact to show a steering group, and the effect arrives as an absence — the meeting that stopped being necessary, the approval nobody reinstated, the objection that came up in week two rather than month six.

That is genuinely difficult if you have to justify your time. It also means the work gets done by people who are prepared to do something valuable without being able to prove they did it, which is a real constraint and worth being honest about rather than motivational.

What I would hold onto is the arithmetic from the observation: these conditions convert into structure either way. Present, they let an organization run on less apparatus than its size suggests. Absent, they quietly commission the apparatus instead — and the apparatus, once built, is nobody’s to remove.

This cultivates: Nobody Writes Down the Cost of Not Trusting, in Part II.