Part III — Cultivating Flow · Chapter 18 · First Public Draft

Growing the Team

Cultivating the conditions where good work keeps happening.

Creates valueOwnership clarity

The cultivation idea first came up in a conversation with my friend and colleague Lars Barkman. We kept turning it over, and the longer we did, the clearer something became that I’d spent years circling without ever saying plainly:

The work isn’t designing solutions. It’s cultivating the conditions where good solutions keep emerging.

That took a while to admit, because designing solutions is the part of the job that feels like the job. It’s visible. It produces something you can present. Cultivating conditions produces, on a good day, nothing you can point at — just an organization that turned out not to need you in the room.

And it isn’t a metaphor. It’s the thing the metaphor had been pointing at all along.

You cannot specify a tree into existence, hand it a deadline, and expect fruit by Friday. You can only create the conditions — soil, light, water, room — and then do the patient, unglamorous work of tending while the tree does what trees do.

Teams are the same. You cannot manage a team into greatness. You can only cultivate the conditions that make greatness possible, and get out of the way often enough to let it happen.

That distinction — between managing and cultivating — is most of what this chapter is about.

This isn’t a software observation. An operating theatre team, a newsroom on deadline, a restaurant kitchen at service — each is a small group with a shared outcome, real interdependence, and no time to route decisions upward. They work well or badly for almost exactly the reasons software teams do, which is a strong hint that the pattern isn’t about the work at all.

A healthy root system is never uniform. Some roots drive deep for water, others spread wide and shallow for what only the topsoil holds, and the fine ones nobody ever sees do most of the actual work. A team works the same way: different competences reaching into different parts of the problem, all drawing on the same ground. The best moments happen when that variety moves in sync — covering for each other, playing to each other’s strengths, growing toward the same light.

Nothing about cultivating is passive, though. Anyone tending a tree still has a plan — which limbs to let run, which to cut back, what needs staking before it splits under its own weight. Teams need the equivalent: guiding principles that connect each day’s work to something larger, so a sprint isn’t just busy, it’s aimed.

Without that plan you don’t get a strong tree. You get scrub.

Underneath all of it is soil, and nothing grows in soil compacted by fear. Trust improves the way soil does — slowly, through repeated seasons of people knowing what to expect from each other. Clarity about vision, goals and responsibility is what keeps the ground loose enough for roots to spread. A culture of open communication and tolerated risk isn’t something you add on top of trust. It is how trust gets built, one honest conversation at a time.

Three conditions show up again and again in Daniel Pink’s work on what makes people thrive — autonomy, mastery and purpose — and they map onto cultivation almost too neatly. Autonomy is room to grow in your own direction within the ground you’ve been given. Mastery is the slow thickening of a root system that only comes from staying in one place long enough to actually take hold. Purpose is what the whole tree is reaching for.

Nobody plants without deciding where one stand ends and the next begins — not out of suspicion, but because clear edges are what let each stand be tended on its own terms, without one plot’s problems choking the plot beside it.

Organizing teams around areas of the business is the same instinct. And how you draw those boundaries matters more than most organizations realize, because the shape of the teams becomes the shape of what they produce, whether anyone planned for it or not. Rigid, centralized communication produces rigid, centralized results. Teams that can act independently produce things that can change independently.

The layout and the yield are never really two separate things — which is the introduction’s arithmetic restated at the scale where you can actually do something about it. You don’t get an adaptable result out of an organization that has to convene to agree on anything.

The healthiest stands, in the end, aren’t the largest. They’re the ones sized so one person can reach every tree without trampling three others to get there. Teams work the same way: small enough that everyone can still see the whole plot, organized around a domain they actually own, given autonomy, mastery and purpose, and cultivated with clarity and trust rather than instruction.

The friction this removes has a specific shape: the waiting, the checking, the asking that fills the space where conditions were never shaped. A team that has to ask permission for every judgement call hasn’t been undertrained. It’s been under-cultivated — nobody ever settled the ground it was meant to grow in.

None of which tells a team what it is for. Conditions make good work possible; they don’t say what work. And the answer to that has to be something more durable than the project currently in front of them, or the conditions will be rebuilt from scratch every time the project ends.

A system that needs cultivating, and a way of cultivating it. That’s the baseline underneath everything this paper calls Organizational Flow.

Who you hire is part of the cultivation

Conditions are not only shaped by what you settle. They are shaped by who is standing inside them — which organizations almost never look at this way, because hiring is filed under recruitment and friction is filed under process, and the two folders are rarely opened in the same meeting.

The advice everyone gives is hire the best people. It sounds unarguable and it is nearly useless, because it contains no information about your organization. Best at what, arranged how, deciding what on their own authority.

I once helped recruit someone genuinely excellent into a structure that could not use them. Entrepreneurial, fast, used to deciding and then telling people afterwards — exactly the profile the hiring brief asked for, and I remember being pleased with the appointment. They arrived into an organization where significant decisions went to a forum that met fortnightly, and where deciding first and explaining later was read as a governance problem rather than as initiative. Within six months they had learned to wait. Within a year they had gone, and the exit conversation was polite and entirely about pace.

Nobody had done anything wrong. The person was excellent, the structure was coherent, and the combination produced friction that neither party had signed up for. What was missing was anyone asking whether the organization we actually had could use the person we were actually hiring.

So the real question is not who is best. It is who will move well through the organization you actually have — or through the one you have decided to build, if you have decided.

The failure runs in both directions, and both are common.

Entrepreneurial people hired into a control structure. They decide, get corrected, decide again, get corrected again, and then stop. What follows is worse than if you had hired someone compliant, because you have paid a premium for judgement and then trained it out. The organization concludes that the hire did not work out. The hire concludes the organization was not serious, and both are describing the same structural mismatch from different sides.

People uncomfortable with responsibility hired into distributed ownership. This one is quieter and takes longer to see. Ownership gets assigned and nothing is decided. Questions travel upward that were meant to stop where they started. Leaders conclude that autonomy does not work here, and quietly recentralize — and the recentralization is now supported by evidence, which makes it very hard to argue with. What actually happened is that the organization staffed a structure with people who never wanted it.

Neither profile is better. Both are entirely reasonable people who would thrive somewhere else in the same market.

Which makes hiring one of the slowest forms of friction to undo. A team boundary can be redrawn in an afternoon. A decision right can be moved with a memo and some follow-through. A mismatch between a person and the structure they work inside lasts as long as they stay, generates friction every week of it, and is nobody’s to fix — because it does not look like friction. It looks like a person.

Three things follow, and none of them require a programme.

Decide the organization first, then hire into it. If you want distributed ownership, say so in the room, and hire people who visibly want to own something. If you want centralized control — which is a legitimate choice for some businesses, and the paper’s argument does not make it illegitimate — then hire people who work well inside it, and stop recruiting on the promise of an autonomy you do not intend to grant.

Interview for the conditions, not only for the craft. The most useful question I know is what someone did the last time they disagreed with a decision that had already been made. The answer tells you where they will sit in your structure far more reliably than a technical assessment does.

Be honest in the recruiting conversation about how decisions actually get made here. Not how the operating model says they get made. The honest version costs you some candidates and saves you the ones who would have left in a year anyway.

The organizations that get this right are not the ones with the strongest employer brand. They are the ones that knew what they were building and hired people who wanted to build that.

Cultivating a team that learns

A team that cultivates well still has to close its own loop, and that is a practice rather than a structure. Small things, and they are small deliberately.

Send the outcome back to the people who chose. Whoever decided to build it hears, six months later, what happened — from someone outside the team, in numbers or in a sentence a customer said. This is a routing problem, not a cultural one, and it is usually solved by one person deciding to forward something.

Give one person the lesson, not the list. A named owner for the one change that follows, with a date. Not fifteen actions. One, owned, done.

Say what would have to be true for the decision to have been right. Written down at the time, before the outcome is known. It converts the later conversation from a judgement about a person into a comparison against something they wrote themselves, which is the cheapest way I know to remove the defensiveness that keeps the loop open.

Let the loop close on things that went well, too. Organizations examine failures and celebrate successes, which means half the available learning is spent on applause. The successes are where the reusable knowledge is, because they can be repeated.

Great teams are grown, not assembled. They need space, light and time — and someone patient enough to keep tending them after the planting is done.

Which is the pattern for everything that follows. Each chapter in this part takes one of the conditions Part II described and asks the same question: not what it is, but how it is deliberately strengthened, by someone, over time.