Return-to-office edicts meet Slack-first habits. Teams organize by channels until deadlines bite, then everyone asks who can approve the trade-off. The fight over hierarchy versus networks is back, yet it misses the point. The power that matters is the right to decide and to enforce the decision.
What the research actually says
Michael Lee and Amy Edmondson offer the clean definition of a self-managing organization: a firm that decentralizes authority formally and systematically, eliminating the manager–subordinate tie as the unit of control (Lee and Edmondson, 2017). Their review of decades of management research draws a boundary line that leaders should heed. Managerial hierarchy is good at reliable execution of known tasks in stable conditions. It is far weaker when the job is complex, non-routine, and spills across functions. There, hierarchy tends to rigidify and to amplify status gaps that stunt people’s development (Lee and Edmondson, 2017). This is not experimental proof, it is a peer-reviewed synthesis, but it is the best map we have of where networked, role-based designs help and where they do not.
Now the counterweight. Nicolai Foss and Peter Klein argue that flat or bossless structures struggle when conditions are dynamic and time-sensitive. Innovation at scale and speed tends to suffer, and many firms that claim flatness quietly retain real bosses, only informal or renamed (Foss and Klein, 2022). This is not an abstract warning. In practice, formal hierarchy often gets replaced by an informal one. Authority does not vanish, it hides.
Seen together, the message is simple. If your work is complex and cross-boundary, rigid ranks get in the way. If your environment is volatile and decisions must be made quickly, blurred authority is a tax on speed. Either way, the variable to manage is decision rights, not the org chart aesthetics.
Two cases, one lesson
Zappos tried to live the no-boss ideal under holacracy. In March 2015 the company’s CEO offered an ultimatum to commit fully or leave. Eighteen percent took a buyout (Groth, 2020). By 2016, Zappos was the largest company to attempt holacracy in full, which is why so many observers watched it as a bellwether (Bernstein, Bunch, Canner and Lee, 2016). By 2017 the firm had moved on. It dropped holacracy’s bureaucratic meetings, brought managers back, and installed an internal market where teams operated like small businesses with their own profit and loss and priced services to one another (Groth, 2020). The form changed, the need did not. Someone still had to decide, allocate, and enforce. Zappos made those rights legible again through managers and prices.
Valve, the games company famed for flatness, shows the other edge. Employees described the firm’s creative state by 2013 as “The Wilderness,” a stall that required management intervention years later to ship Half-Life: Alyx despite the bossless ideal (Foss and Klein, 2022). Commitment to flatness also meant no dedicated customer support and marketing departments, a choice with clear costs when products meet users at scale (Foss and Klein, 2022). Again, authority did not disappear. It reappeared when delays and missing capabilities made it impossible to hide.
These cases do not prove that hierarchy is superior. They show that when decision rights are vague, organizations eventually improvise a hierarchy to move. The improvisation is slow and political. The fix is to make the rights explicit before the crunch.
The real variable: who decides, how, and with what teeth
A balanced reading of the bossless debate concludes that there is no universally superior structure. Near-bossless firms are rare and do not show across-the-board performance advantages over conventional ones, and managers matter for concrete functions: to design structures and processes, organize work, encourage action, and enforce decisions (strategy+business, 2022). Foss and Klein distinguish between hard authority to hire, fire, and instruct, and softer authority to structure, lead, and build consensus. Both types exist in effective firms, and which to lean on is contingent on context (strategy+business, 2022).
Now apply that to the earlier research. Self-managing designs decentralize authority in ways that help cross-functional problem solving, but only if people know who can commit resources and who arbitrates conflicts across roles and teams (Lee and Edmondson, 2017). Flat structures that refuse to name those rights end up with shadow bosses, often the most senior or forceful personalities, which is the worst of both worlds: politics without clarity (Foss and Klein, 2022).
A practical playbook for power you can execute
Most organizations do not need a doctrinal shift to self-management or a reversion to old hierarchies. They need legible decision rights. Four moves help:
-
Map decisions, not titles. List the recurring, consequential calls in your business: who spends money above a threshold, who can ship a product change, who can stop a launch, who sets prices. Publish the names or roles. Treat this as living policy. This turns informal authority into shared clarity.
-
Predefine escalation for time-sensitive work. When issues cross team boundaries or deadlines loom, assign a temporary decision owner with a clock and a mandate to choose. The research warning is clear: dynamic, time-sensitive environments punish ambiguity (Foss and Klein, 2022).
-
Make enforcement visible. State who can say no and on what grounds. Managers are not ornaments. They exist to design, organize, encourage, and enforce, especially when consensus stalls (strategy+business, 2022).
-
Prune governance that does not change decisions. Zappos stripped out holacracy rituals when they became overhead and restored managers where accountability had thinned (Groth, 2020). Review your meetings and processes with the same lens. If a ritual never changes an outcome, cut it or redesign it.
The upside of this approach is not ideological purity. It is speed with accountability. In a year of budget cuts and large bets on AI, boards should ask to see decision rights, not seat charts. Bosses and networks are tools. Use both, but design the power that moves work from talk to done.
Sources
- Lee and Edmondson (2017). Self-managing organizations: Exploring the limits of less-hierarchical organizing. Defines SMOs and synthesizes when hierarchy aids reliability versus when it hinders complex, cross-boundary work, grounding the argument for context-contingent authority. https://www.sciencedirect.com/science/article/abs/pii/S0191308517300059
- Foss and Klein (2022). Who's the boss? The persistence of entrepreneurial hierarchy in flat organizations. Shows that flat firms often develop informal hierarchies and struggle in dynamic, time-sensitive settings, supporting the case for explicit decision rights. https://link.springer.com/article/10.1007/s41469-022-00132-8
- Groth (2020). Zappos has quietly backed away from holacracy. Documents Zappos’s 2015 ultimatum and 18% buyout, and its 2017 shift to an internal market with managers returning, illustrating governance reversals when decision rights are unclear. https://qz.com/work/1776841/zappos-has-quietly-backed-away-from-holacracy
- Bernstein, Bunch, Canner and Lee (2016). Beyond the Holacracy Hype. Notes that by 2016 Zappos was the largest firm attempting holacracy and argues for a balanced assessment of self-management, providing context for the polarized debate. https://hbr.org/2016/07/beyond-the-holacracy-hype
- strategy+business (2022). Can bossless management work? Summarizes Foss and Klein’s view that near-bossless firms are rare, lack proven superior performance, and that managers matter through design, organization, encouragement, and enforcement, reinforcing the practical role of authority. https://www.strategy-business.com/article/Can-bossless-management-work