Every wealthy country is short of nurses and heavy with administration. When care quality slips, the standard reflex is to add a layer: another supervisor, another compliance officer, another manager to watch the managers. One Dutch home care organization spent nearly two decades doing the reverse, and the model is now documented carefully enough to argue about seriously rather than admire from a distance.
A structure built by subtraction
Buurtzorg, founded by the former nurse Jos de Blok, delivers community nursing through small, self-directed teams. In the most rigorous account of the model to date, Martela and Nandram (2025) describe an organization running more than 900 self-managing teams, each capped at roughly twelve nurses, employing about 14,700 people as of January 2021, and doing it with only two directors, around fifty back-office staff, and no middle managers or supervisors at all. Teams set their own rosters, decide whom to hire, hold their own budgets, and coordinate directly with patients and local doctors. Nobody signs off on their daily decisions because there is nobody positioned above them to do so.
The figure that unsettles most executives is the support ratio. Martela and Nandram (2025) report that the whole network of roughly 950 teams was served by just 21 internal coaches, better than one coach for every 45 teams, and that those coaches held no formal authority and bore no responsibility for team productivity. They advise when asked. They cannot command. Set that against the supervisory scaffolding of a conventional provider, where spans of control of six or eight are considered generous, and the difference is not incremental. It is structural. The organization did not flatten a pyramid so much as decline to build one.
The coaches are worth dwelling on, because they are the part outsiders most often misunderstand. A coach in this system cannot rescue a struggling team by taking the wheel, since doing so would recreate the very supervisor the model was built to avoid. Martela and Nandram (2025) describe a role confined to guidance and reflection, which is why a single coach can plausibly stay in touch with dozens of teams at once. If those coaches carried targets and accountability for output, the arithmetic would break instantly, and the organization would need the layer of managers it has spent two decades doing without.
What the model actually rests on
The temptation is to read this as proof that hierarchy is waste. Martela and Nandram (2025) are more careful, and so should anyone borrowing the model be. They identify explicit boundary conditions: the approach works best where interdependence between teams is low, where the frontline is highly motivated and genuinely competent, and where the service itself lends to local autonomy. Home nursing fits almost perfectly. Each team serves its own neighborhood, the decisive clinical knowledge sits with the nurse at the bedside, and one team's caseload rarely blocks another's. Change those conditions, put the work in tightly coupled production or in the hands of a demoralized frontline, and the same design would strain.
Two caveats keep the story honest. The widely repeated claim that Buurtzorg's overhead ran near 8 percent against a sector average around 25 percent traces to a KPMG assessment from roughly a decade ago, so it describes a past snapshot rather than a permanent law. And the model is not cheap by the hour. Martela and Nandram (2025) note that Buurtzorg's cost per hour has actually run higher than a comparison benchmark, around 54 euros against roughly 49. The savings, where they exist, come from needing fewer hours and producing better outcomes, not from underpaying nurses or thinning the service. Any hospital executive who sells self-management as a labor discount has misread the evidence before implementing it.
A map, not a manifesto
Buurtzorg is often invoked as the proof that bossless organizations work, and the network around it keeps that reputation alive. Corporate Rebels (2025) cite the company, with its roughly 15,000 nurses in small autonomous teams, as the health care organization that revolutionized community care in the Netherlands, and place it at the center of a wider movement of self-managed firms. That reputation is earned. It is also easy to misapply.
The useful comparison is Morning Star, the American tomato processor that Hamel (2011) documented running without managers for more than two decades, where compensation is set by peers and there are no titles or promotions. Two firms, two continents, two industries, and yet the underlying move is identical: authority is handed to the people closest to the work, and the organization invests in the coordination that lets them use it well. What differs is everything local. Morning Star leans on written peer agreements between colleagues. Buurtzorg leans on small neighborhood teams and a thin ring of coaches. Neither imported the other's scaffolding, because scaffolding is where self-management succeeds or fails.
For health systems staring at burnout and vacancy rates, the lesson is not to delete the org chart on Monday. It is to ask whether the conditions Martela and Nandram (2025) name are present: loose coupling between units, expert and motivated staff, and work that genuinely benefits from local judgment. Where they hold, the supervisory layer is doing less than its cost implies, and twenty-one coaches can carry an organization of fourteen thousand. Where they do not, bolting coaches onto a system built for control just produces confusion with a friendlier title. The map tells you where the model travels. It does not promise the territory everywhere.
Sources
- Martela, F. and Nandram, S. (2025). Buurtzorg: scaling up an organization with hundreds of self-managing teams but no middle managers. Peer-reviewed case documenting 900+ teams, ~14,700 staff, two directors and 21 coaches with no formal authority, plus the boundary conditions and cost caveats the article relies on. https://link.springer.com/article/10.1007/s41469-024-00184-y
- Hamel, G. (2011). First, Let's Fire All the Managers. Documents Morning Star running for two decades without managers and with peer-set pay, the American counterpart used for contrast. https://hbr.org/2011/12/first-lets-fire-all-the-managers
- Corporate Rebels (2025). Big news: Frederic Laloux, Jos de Blok, Lisa Gill at Corporate Rebels Summit 2025. Community source citing Buurtzorg's ~15,000 nurses and its role in Dutch community care, used for reputation context, not numbers. https://www.corporate-rebels.com/blog/big-news-frederic-laloux-jos-de-blok-lisa-gill-at-corporate-rebels-summit-2025