Return-to-office orders are back, along with layers of approvals meant to restore control. At the same time, employees bristle at managerial drag and the slow walk of decisions that cross silos. The holacracy hype cycle has ended, not least after high-profile reversals. What survives is a harder question: can self-management deliver in the 2020s without fantasy or free fall?
What self-management really is
Self-management is often caricatured as a company with no bosses. The serious version is more demanding. In their canonical review, Lee and Edmondson (2017) define self-managing organizations as those that decentralize authority radically, formally and systematically across the whole firm, and that eliminate the manager–subordinate relationship as the core organizing principle. Their synthesis of decades of management research draws a useful line. Traditional hierarchy is well suited to reliable execution of known tasks in stable environments. It struggles when work is complex, non-routine, and crosses functional boundaries, where status differences and rigid structures dampen learning and problem solving (Lee and Edmondson, 2017).
That framing matters now. Hybrid schedules and AI tooling have turned many jobs into cross-boundary puzzles. The choice is not between control and chaos, but between a design that clarifies who decides what at the edge, and one that pulls everything back to the center the moment uncertainty rises. Self-management, properly built, is the former.
What the people data actually say
Critics often warn that removing layers will simply push stress down to teams. Recent measurement work complicates that view. Using two independent samples, one of employees across multiple organizations (N = 425) and a nationally representative Finnish sample (N = 2,000), Morikawa, Martela and Hakanen (2025) developed validated scales for two distinct constructs: organizational self-management and individual self-management. Both were positively related to work engagement and negatively related to burnout. The authors also found that job demands partially mediated these relationships, indicating more nuanced pathways rather than a simple “no bosses equals less stress” story (Morikawa, Martela and Hakanen, 2025).
These are correlations, not causal proofs. They do not mean any firm that flattens will automatically boost engagement or cut burnout. They do mean that, in the wild, greater self-management at the organizational and individual levels is associated with healthier people outcomes, and that the old intuition that structure is the only antidote to overload is too narrow. For leaders, the practical read is to treat team autonomy and clearer local decision rights as ingredients in a well-being strategy, then measure effects with credible tools rather than anecdotes.
Scaling without middle managers: the Buurtzorg case
Most executives accept that small teams can self-govern. The fear is scale. Buurtzorg, the Dutch home-care organization, is the most documented counterexample. As of January 2021 it operated more than 900 self-managing teams, typically up to 12 nurses per team, with roughly 14,700 employees, and no middle managers or supervisors. The central staff was lean: only two directors and about 50 people in back-office roles. A network of 21 internal coaches supported about 950 teams, roughly one coach for more than 45 teams, and those coaches had no formal authority and no responsibility for team productivity (Martela and Nandram, 2025).
The economics are subtler than folklore suggests. A KPMG analysis from around 2015 put Buurtzorg’s overhead at about 8 percent versus a 25 percent sector average in Dutch home care. That figure is dated and should not be read as a current benchmark. Moreover, the cost per hour was higher, €54.47 versus €48.74, which tempers any claim that self-management is simply cheaper by design (Martela and Nandram, 2025). Cost structures shift with context. What stands out is the operating pattern: decision rights at the edge, minimal central staffing, and development invested in coaching rather than supervision.
Boundaries matter. Martela and Nandram (2025) identify conditions under which the model works best: relatively low interdependence between teams, highly motivated and competent frontline professionals, and services that let small units handle most decisions locally. Those are design constraints, not footnotes. They point to where to pilot and what not to promise. In domains with dense interdependencies or unstable interfaces, trying to run hundreds of teams without an integrating layer will fail fast. Where the work is modular and expertise-driven, Buurtzorg shows that scale without middle managers is not mythology.
The Zappos detour and a better question
If Buurtzorg is the quiet case for patient design, Zappos is the loud case for overreach. In 2016 Zappos was the largest firm to have implemented holacracy across the board, which helped make it a totem in the debate (Bernstein et al., 2016). Then came the backlash. In March 2015, CEO Tony Hsieh issued an ultimatum requiring full commitment to holacracy or exit, and 18 percent of employees took a buyout. By March 2017 the company had replaced holacracy with an internal market model in which teams operated as small businesses with their own P&Ls and traded with each other at market-like prices. Managers returned, and the firm shed the bureaucratic meeting cadences of holacracy, keeping only a circular hierarchy as a vestige. Leaders cited “big challenges” in business metrics and the critique that holacracy was excessively inward focused (Groth, 2020).
Two lessons follow. First, rhetoric about bossless utopias invites false choices. The serious design question is where authority should live, how to make that visible, and how to help teams navigate trade-offs without bottlenecking on managers. Second, governance is not ideology. Zappos’ shift from holacracy to an internal market shows that firms can and should rework operating systems when business performance demands it, rather than defend a brand of self-organization for its own sake.
The debate itself has been polarized for a decade, with cheerleaders and detractors talking past each other. Bernstein et al. (2016) argued for a more balanced assessment that looks at when and how self-organization helps, not whether it flatters a philosophy. That counsel has aged well. The evidence base is still mostly cases and correlations, but it is specific enough to guide real choices.
The practical read
Leaders facing hybrid coordination problems and tighter budgets do not need a leap of faith. They need a testable design. Start where interdependence is modest and expertise is high, because that is where decentralized authority is most likely to pay off, a boundary drawn by Martela and Nandram (2025). Make the rules of the game explicit, drawing on Lee and Edmondson’s (2017) emphasis on formal, systematic decentralization rather than informal heroics. Replace layers of supervision with lightweight coaching, not shadow management, and track engagement and burnout with validated instruments rather than sentiment, taking a cue from Morikawa, Martela and Hakanen (2025). Finally, communicate the trade-offs plainly. Overhead can fall, but not everywhere and not always, and unit costs may rise even when outcomes improve, as the Buurtzorg case reminds us. Self-management is not a vibe. It is an operating choice that earns its keep when the structure fits the work.
Sources
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing. Defines self-managing organizations and synthesizes when hierarchy hinders complex, cross-boundary problem solving. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Morikawa, Martela and Hakanen (2025). Self-management and employee well-being: validated scales, engagement and burnout. Finds organizational and individual self-management positively associated with engagement and negatively with burnout, with partial mediation by job demands; correlational evidence that informs design. 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. Documents scale without middle managers, dated overhead comparisons, higher hourly costs, coaching design, and boundary conditions for where the model works. https://link.springer.com/article/10.1007/s41469-024-00184-y
- Bernstein, Bunch, Canner and Lee (2016). Beyond the Holacracy Hype. Argues for a balanced assessment of self-organization and notes Zappos as the largest holacracy adopter at the time, framing the polarized debate. https://hbr.org/2016/07/beyond-the-holacracy-hype
- Groth (2020). Zappos has quietly backed away from holacracy. Reports the 2015 ultimatum and 18% buyout, the 2017 shift to an internal market with managers reintroduced, and the stated business reasons for the change. https://qz.com/work/1776841/zappos-has-quietly-backed-away-from-holacracy