Line up the well-known cases and a shape appears. Community nursing, a tomato processor organized around independent contracts, a bank run through autonomous branches, software teams, professional and creative services. It looks eclectic until you ask what these have in common at the level of the work itself, and then the eclecticism dissolves into a pattern that the research on Buurtzorg names directly.
The peer-reviewed study of Buurtzorg (Martela and Nandram, 2025) is unusually explicit that the model works best under three conditions: low interdependence between teams, so a team can act without constantly waiting on others; frontline workers who are highly motivated and highly competent, so that autonomy produces good decisions; and highly customized services, where local judgment beats a central script. Those three conditions are a fair X-ray of every durable self-management case, not just the Dutch one. Where they hold, self-management thrives. Where they fail, it strains.
The three conditions, and what they explain
Consider low interdependence first. A community nurse can serve her patients whether or not the team across town is having a good week. A branch of Handelsbanken can win and keep its customers without hourly coordination with other branches. Contrast that with a tightly coupled assembly line where one station's timing dictates the next, or a business where every output requires five teams to synchronize continuously. High forced interdependence needs a coordinator, and if you remove the manager without removing the coupling, someone has to reinvent coordination under pressure. Self-management fits loosely coupled work.
Second, decisive frontline judgment. Self-management is a bet that the person closest to the work knows best. That bet pays when the work genuinely rewards expert, local judgment: a nurse reading a patient, an engineer solving a novel fault, a banker sizing up a borrower. It pays less where the work is simple, standardized, and better served by a reliable script than by individual discretion. This is why care, engineering, and craft-heavy manufacturing recur, and why routine, high-volume, low-variance work does not.
Third, customized output. Where every case is a little different, central rules cannot anticipate the variety, and pushing authority down is the only way to match response to reality. Where output is uniform and the premium is on identical repetition, the case for local discretion weakens.
The counterexamples that prove the rule
Two cautions keep this from hardening into a law. The first is that the conditions cut across the obvious sector labels. Manufacturing is usually cited as hostile to self-management, yet Morning Star runs a heavy industrial process without managers, because it engineered low interdependence through explicit peer contracts rather than assuming the assembly line forbids it. Ricardo Semler's Semco made the same point in a different key. As Ricardo Semler has told it, Semco's plants were often run by people without the credentials the org chart would have demanded, including a maintenance technician without a degree leading two hundred people because everyone trusted him, and his rule was that trust does not come with a diploma. The sector was industrial; the conditions were engineered to fit. Sector is a rough proxy. The conditions are the real variable.
The second caution comes from the skeptics. Foss and Klein (Journal of Organization Design, 2022) argue that flat organizations tend to struggle in dynamic, time-sensitive, or highly interdependent settings, and that many firms claiming to be bossless simply retain informal hierarchy under new names. Their examples, such as Valve's drift into what employees called creative inertia, are cases where the three conditions were absent or eroded. That is not a refutation of the pattern; it is confirmation from the failure side. Remove low interdependence or decisive frontline judgment, and self-management does exactly what the theory predicts it will: it wobbles.
So the clustering is not fashion or founder charisma. It is fit. Before adopting self-management, the useful question is not "which sector am I in" but "does my work have low interdependence, decisive frontline judgment, and customized output." Where all three are present, self-management is pushing on an open door, whatever the industry. Where they are absent, no amount of enthusiasm will hold it, and the honest move is either to engineer the conditions on purpose, as Morning Star did, or to leave the hierarchy where the work actually needs it.
Sources
- Martela, F. & Nandram, S. (2025). Buurtzorg: scaling up an organization with hundreds of self-managing teams but no middle managers. Journal of Organization Design 14. Names the boundary conditions for self-management: low inter-team interdependence, highly motivated and competent frontline workers, and highly customized services. https://link.springer.com/article/10.1007/s41469-024-00184-y
- Foss, N. & Klein, P. (2022). Who's the boss? The persistence of entrepreneurial hierarchy in flat organizations. Journal of Organization Design. Flat organizations struggle in dynamic, interdependent settings and often retain informal hierarchy under new names. https://link.springer.com/article/10.1007/s41469-022-00132-8
- Hamel, G. (2011). First, Let's Fire All the Managers. Harvard Business Review. Morning Star shows heavy manufacturing can be self-managed by engineering low interdependence through explicit peer contracts. https://hbr.org/2011/12/first-lets-fire-all-the-managers