Executives want speed without chaos and accountability without bureaucracy. Many have tried to get both by flattening. Zappos offered the most famous test. After a 2015 ultimatum to adopt holacracy or leave, 18 percent of employees took a buyout. Two years later the firm abandoned holacracy’s routines, reinstated managers, and pivoted to an internal market where teams ran with their own P&Ls and traded with one another at market prices (Groth, 2020). The promise was elegant: replace bosses with prices. The reality was starker. Power did not vanish, it moved.
What bossless actually means
Self-managing organizations, in the canonical definition, decentralize authority radically and systematically, eliminating the manager–subordinate relationship as the default unit of control (Lee and Edmondson, 2017). The appeal is not mystical. Managerial hierarchy is effective at reliable execution in stable settings. It is also prone to rigidity, status barriers, and the dampening of cross-boundary problem solving and individual growth (Lee and Edmondson, 2017). That trade-off explains why networked models surface in sectors where novelty and coordination across functions are constant.
The debate around these models has been unhelpfully binary. In 2016, a widely read assessment urged a more balanced view, noting both the allure and the operational friction of bossless systems and observing that Zappos was then the largest firm to adopt holacracy wholesale (Bernstein et al., 2016). The issue is not faith in circles versus cynicism about them. It is whether a given design supplies the authority and coordination a business actually needs, on the clock and at scale.
When the vacuum fills with informal power
Flat firms do not exist in a power vacuum. They tend to grow informal hierarchies or rely on senior figures whose authority is real even if it is not mapped on an org chart. That is not a moral failing, it is an organizational response. In dynamic and time-sensitive environments, near-bossless structures often struggle with innovation at scale and with sustained performance over time (Foss and Klein, 2022). The case of Valve, long held up as a bossless exemplar, is instructive. By 2013 employees described a period of creative drift as “The Wilderness,” and the launch of Half-Life: Alyx required management intervention despite the company’s avowedly flat model. Valve’s reluctance to build conventional customer-support and marketing functions, in service of preserving flatness, limited downstream capabilities the business still needed (Foss and Klein, 2022).
These are not isolated anecdotes. They point to a pattern. Remove formal hierarchy and you often replace it with informal hierarchy, cliques, or charismatic authority. Coordination problems do not solve themselves, they surface elsewhere. The research record and the lived cases counsel sobriety: managers matter not as ornamental titleholders but because someone has to design structures, set priorities, and enforce decisions when consensus stalls (strategy+business, 2022). Authority shows up in two forms in that account. There is the classic power to hire, fire, and instruct, and there is the quieter work of structuring, persuading, and securing agreement across teams. Flattening attempts to mute the first, but it cannot wish away the second. Nor should it.
Zappos’s market turn is a governance redesign, not an escape from power
Zappos’s shift from holacracy to market-based dynamics is often framed as a retreat. It is better seen as a change in governance. The firm cut back on holacracy’s meeting rituals, brought managers back into the system, and asked teams to operate as small businesses trading with one another at internally set prices (Groth, 2020). That move addressed a critique that holacracy had become inwardly focused and operationally heavy. It also put a price on coordination. When teams account for their own P&Ls, resource allocation becomes legible and tradeoffs become explicit. But pricing does not resolve questions of direction or timing. Internal markets need rules, dispute resolution, and the equivalent of a central bank when shocks hit. Someone still sets the protocols and adjudicates exceptions.
The broader record reinforces this point. Reviews of near-bossless firms conclude that these designs do not consistently outperform traditional ones in the aggregate, and that there are relatively few large, durable examples to study. The same reviews emphasize the practical value of managers who design, organize, encourage, and enforce, and who distinguish between hard-line authority and the authority to structure and build consensus (strategy+business, 2022). These are governance levers. Markets inside the firm can complement them, not replace them.
A design brief for networks that face the clock
Three implications follow from the evidence and the cases.
- Make authority explicit, even when it is distributed. Define who decides what, on what criteria, and by when. Self-managing systems work only when the elimination of the manager–subordinate tie is matched by clear, formal decision rights elsewhere (Lee and Edmondson, 2017).
- Expect informal hierarchy and channel it, rather than pretending it will not appear. Identify the senior voices people already follow and give them bounded, transparent mandates tied to specific outcomes and time frames. The alternative is shadow power that evades accountability, the pattern that trips up many “flat” firms (Foss and Klein, 2022).
- Use internal prices as signals, not as a substitute for leadership. Team P&Ls and internal chargebacks can clarify costs and surface tradeoffs, as Zappos sought to do, but they do not settle strategic priorities or resolve stalemates. Keep a visible escalation path and time-boxed roles that can break ties when markets misfire (Groth, 2020; strategy+business, 2022).
The practical question is not whether hierarchy or network “wins.” It is whether your structure reveals power and lets people use it responsibly under pressure. The academic synthesis reminds us why we bother: hierarchy is strong medicine for reliability in stable work, but it can suppress the cross-cutting problem solving that complex work demands (Lee and Edmondson, 2017). The contemporary cases remind us of the bill: bossless rhetoric often decays into informal rule, and internal markets require governance to function at scale (Foss and Klein, 2022; Groth, 2020).
Leaders should redesign, not romanticize. If you want the speed of a network and the dependability of a line operation, put prices where they help, publish decision rights, and give authority a face when the clock is ticking. Power will exist either way. Your choice is whether it is visible and accountable, or hidden and capricious.
Sources
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing. Defines SMOs and outlines when hierarchy helps reliability and when it stifles cross-boundary problem solving, grounding the trade-offs discussed. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Bernstein, Bunch, Canner and Lee (2016). Beyond the Holacracy Hype, Harvard Business Review. Argues for a balanced assessment of bossless systems and notes Zappos as the largest holacracy adopter at the time, framing the debate. https://hbr.org/2016/07/beyond-the-holacracy-hype
- Foss and Klein (2022). Who's the boss? The persistence of entrepreneurial hierarchy in flat organizations, Journal of Organization Design. Shows that flat firms often develop informal hierarchies and illustrates limits through Valve’s “Wilderness” period and capability gaps. https://link.springer.com/article/10.1007/s41469-022-00132-8
- Groth (2020). Zappos has quietly backed away from holacracy, Quartz. Reports the 2015 ultimatum with an 18% buyout and the 2017 shift to an internal market with managers reinstated, clarifying what changed in governance. https://qz.com/work/1776841/zappos-has-quietly-backed-away-from-holacracy
- strategy+business (2022). Can bossless management work? Summarizes Foss and Klein’s view that near-bossless firms do not reliably outperform and that managers matter through distinct forms of authority. https://www.strategy-business.com/article/Can-bossless-management-work