Return-to-office mandates, cost pressure, and morale dips have revived an old temptation: flatten the org and trim the middle. Some boards now flirt with self-management as a shortcut to agility and savings. The question is not whether you can remove bosses. It is whether you have built the mechanisms that make bosses redundant.
What self-management actually changes
Self-management is a specific design choice, not a vibe. Michael Y. Lee and Amy C. Edmondson define self-managing organizations as those that decentralize authority radically, formally, and systematically, and whose distinguishing feature is the elimination of the manager–subordinate relationship across the enterprise (Lee and Edmondson, 2017). Their review of the management literature is clear on the trade: traditional hierarchy excels at reliable execution of known tasks in stable settings, yet it tends to suppress complex, cross-functional problem solving, and it hardens status differences that stall people’s development (Lee and Edmondson, 2017). Self-management attempts to invert that equation by moving decisions closer to the work and making roles, not ranks, the unit of authority.
That shift replaces a person with a fabric of rules, forums, and peer commitments. It is a heavier lift than it sounds. The work managers used to do does not vanish. Someone must still coordinate, set standards, allocate resources, and handle exceptions. In successful cases those functions are redistributed into clear processes, transparent metrics, and light-touch coaching, then enforced by norms and contracts rather than by positional power.
The case that it can scale
Buurtzorg shows that this replacement can work at industrial scale, in a domain most leaders would consider managerial by default: community nursing. By January 2021 the Dutch firm was operating more than 900 self-managing teams of up to 12 nurses, employing about 14,700 people, with only two directors, about 50 people in the back office, and no middle managers or supervisors (Martela and Nandram, 2025). The backbone is a sparse layer of 21 internal coaches who support roughly 950 teams. These coaches carry no formal authority and are not accountable for team productivity (Martela and Nandram, 2025). That choice is not decorative. It forces the work of coordination back into team routines and peer agreements.
The economics are not a fairy tale. Reported overhead was 8 percent compared with a 25 percent sector average in Dutch home care, according to a circa 2015 KPMG analysis cited in the case, a dated benchmark that should not be presented as current (Martela and Nandram, 2025). At the same time, Buurtzorg’s cost per hour was higher than peers (€54.47 versus €48.74), an important offset that keeps the story grounded (Martela and Nandram, 2025). The authors identify boundary conditions as well. The model performs best when interdependence between teams is low and frontline staff are both motivated and competent, and it is tailored to specific service contexts rather than every kind of work (Martela and Nandram, 2025). In other words, self-management is a fit problem before it is a faith statement.
What the well-being data actually say
A common worry is that removing managers simply transfers stress to employees. A recent multi-sample study offers a more nuanced picture. Using validated scales in a sample of 425 employees across organizations and a nationally representative sample of 2,000 in Finland, researchers found that both organizational self-management and employee self-management are positively associated with work engagement and negatively associated with burnout (Morikawa, Martela and Hakanen, 2025). The relationships are correlational rather than causal. Even so, the findings counter the idea that less hierarchy inevitably breeds overload. The authors also report partial mediation through job demands, which suggests that how work is designed and distributed under self-management matters for well-being outcomes (Morikawa, Martela and Hakanen, 2025).
Leaders should read this as a design signal. Self-management is not a free-choice festival. It requires tightening the system where it is loose and loosening it where it is tight. Clear decision rights, visible workloads, and codified ways to resolve conflicts limit unnecessary demands and amplify the autonomy that drives engagement.
The hype trap, and its tuition bill
The public debate about “bossless” companies has been polarized for a decade. Ethan Bernstein and coauthors argued for a more balanced view in 2016, noting both the promise and the pitfalls as experiments scaled, and they reported that Zappos was then the largest firm to adopt Holacracy (Bernstein, Bunch, Canner and Lee, 2016). The subsequent arc at Zappos is instructive. In March 2015 the CEO gave employees a stark choice: fully embrace Holacracy or take a buyout. Eighteen percent took the exit (Groth, 2020). By 2017 the company had shifted to an internal market model in which teams operated like small businesses with their own P&Ls, and it reintroduced managers while shedding the more bureaucratic Holacracy rituals. The circular hierarchy remained mostly as an artifact (Groth, 2020).
This is not a morality tale about self-management failing. It is a case about underestimating the operating burden of the mechanisms that substitute for managers, and about the costs of adopting a totalizing system unsuited to the business. Market-based dynamics brought their own explicit rules of coordination, pricing, and accountability. The firm moved from one set of mechanisms to another, and the shift came only after a painful reckoning with process complexity and performance concerns (Groth, 2020).
The practical read
If you plan to flatten, budget for the scaffolding. The Buurtzorg case points to a lean coaching layer with no formal authority, disciplined team sizes, and a back office sized for enablement rather than control (Martela and Nandram, 2025). The well-being evidence suggests that autonomy without guardrails does not deliver. Design the work so that decision rights are unambiguous, cross-team interdependencies are minimal or explicitly managed, and conflict resolution does not rely on heroics (Morikawa, Martela and Hakanen, 2025). Be explicit about where the model fits best. Lee and Edmondson’s synthesis is blunt on context: hierarchy is effective for reliable execution in stable environments, while self-management aims at complex, nonroutine work that cuts across functions (Lee and Edmondson, 2017).
Finally, treat frameworks as toolkits, not dogma. The Zappos journey shows that governance models are means to an end and that changing mechanisms midstream is costly in attention and attrition (Bernstein, Bunch, Canner and Lee, 2016; Groth, 2020). Start where the dependencies are simplest and the professionals most prepared to own outcomes. Then scale what holds under scrutiny. Replace bosses with mechanisms. Replace slogans with design.
Sources
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing. Defines self-management and clarifies where hierarchy excels versus where decentralization aims to outperform, grounding the argument about fit. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Bernstein, Bunch, Canner and Lee (2016). Beyond the Holacracy Hype. Describes the polarized debate and notes Zappos as the largest Holacracy adopter at the time, supporting the caution against ideology. https://hbr.org/2016/07/beyond-the-holacracy-hype
- Morikawa, Martela and Hakanen (2025). Self-management and employee well-being: validated scales, engagement and burnout. Finds positive associations with engagement and negative with burnout, with correlational limits, informing the well-being section. https://journals.sagepub.com/doi/10.1177/23409444241307734
- Martela and Nandram (2025). Buurtzorg: scaling up an organization with hundreds of self-managing teams but no middle managers. Details the structure, coaching model, dated overhead comparison, higher hourly costs, and boundary conditions, illustrating scalable mechanisms and their limits. https://link.springer.com/article/10.1007/s41469-024-00184-y
- Groth (2020). Quartz — Zappos has quietly backed away from holacracy. Reports the 2015 buyout accepted by 18 percent, the 2017 pivot to an internal market model, and the reintroduction of managers, a cautionary case about adopting and revising self-management systems. https://qz.com/work/1776841/zappos-has-quietly-backed-away-from-holacracy