What makes self-managing organizations novel? Comparing how Weberian bureaucracy, Mintzberg’s adhocracy, and self-organizing solve six fundamental problems of organizing

Abstract

The bureaucratic organizational structure has been recently challenged by a number of organizations that claim to offer employee emancipation and autonomy through self-management, self-organizing, or “holacracy.” To facilitate theorizing about such organizational-level self-management, I examine it as an ideal type of organizational form, comparing it to two more established organizational forms, Weberian bureaucracy and Mintzberg’s adhocracy. More particularly, building on the four universal problems every organization needs to solve—two of which I divide into two sub-problems—I utilize a framework of six fundamental problems of organizing—task division, task allocation, rewarding desired behavior, eliminating freeriding, providing direction, and ensuring coordination—to demonstrate how these three forms of organizing have found different solutions to them. The radically decentralized model of authority at the heart of self-managing organizations is shown to lead to solutions to these problems that are based on peer-based accountability and rewarding, transparency of key information, and bottom-up emergent processes where employees have the authority and responsibility to identify necessary tasks and ensure that they get done. It is concluded that the self-managing organization indeed is a novel form of organizing that can better explain certain real-life organizational outliers than the existing paradigms of organizing. It is argued to be especially viable in industries where interdependence between units is low, outputs are highly tailored, and employee expertise and motivation are high. Accordingly, research on such organizations can offer several new insights relevant to both the practice and theory of organization design.

Full text

R E S E A R C H Open Access
What makes self-managing organizations
novel? Comparing how Weberian
bureaucracy, Mintzberg’s adhocracy, and
self-organizing solve six fundamental
problems of organizing
Frank Martela
Correspondence: frank.martela@
gmail.com
Aalto University, P.O. Box 15500,
00076 Aalto, Finland
Abstract
The bureaucratic organizational structure has been recently challenged by a number
of organizations that claim to offer employee emancipation and autonomy through
self-management, self-organizing, or “holacracy.”To facilitate theorizing about such
organizational-level self-management, I examine it as an ideal type of organizational
form, comparing it to two more established organizational forms, Weberian
bureaucracy and Mintzberg’s adhocracy. More particularly, building on the four
universal problems every organization needs to solve—two of which I divide into
two sub-problems—I utilize a framework of six fundamental problems of
organizing—task division, task allocation, rewarding desired behavior, eliminating
freeriding, providing direction, and ensuring coordination—to demonstrate how
these three forms of organizing have found different solutions to them. The radically
decentralized model of authority at the heart of self-managing organizations is
shown to lead to solutions to these problems that are based on peer-based
accountability and rewarding, transparency of key information, and bottom-up
emergent processes where employees have the authority and responsibility to
identify necessary tasks and ensure that they get done. It is concluded that the self-
managing organization indeed is a novel form of organizing that can better explain
certain real-life organizational outliers than the existing paradigms of organizing. It is
argued to be especially viable in industries where interdependence between units is
low, outputs are highly tailored, and employee expertise and motivation are high.
Accordingly, research on such organizations can offer several new insights relevant
to both the practice and theory of organization design.
Keywords: Adhocracy, Bureaucracy, Emancipation, Hierarchy, Motivation,
Organizational design, Organizational theory, Self-management, Self-organizing
Introduction
Recent years have seen an upsurge of academic (Puranam and Håkonsson 2015;
Martela and Kostamo 2017; Burton et al. 2017), influential business review (Hamel
2011; Bernstein et al. 2016), and popular interest (Gelles 2015; Robertson 2016) in self-
managing organizations (SMO), which are defined as organizations that have radically
© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International
License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium,
provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and
indicate if changes were made.
Martela Journal of Organization Design (2019) 8:23
https://doi.org/10.1186/s41469-019-0062-9

and systematically decentralized authority throughout the organization to the degree of
almost abolishing the whole layer of middle management and supervisor-subordinate
relationships (Lee and Edmondson 2017). Zappos, Buurtzorg, Morning Star, and Valve
have arguably been most in the spotlight, but several other examples such as W.L.
Gore, Hayer, and Reaktor also exist. While self-managing at the team level has been
discussed at least since the 1950s (Trist and Bamforth 1951), with an upsurge of inter-
est in the 1980s and 1990s (e.g., Cohen and Ledford 1994; Ezzamel and Willmott
1998), only in recent decades have we started to see examples of organizations with
thousands of employees that are ”applying the principles of self-management to entire
institutions”(Bernstein et al. 2016, p. 42) through abolishing leader-subordinate rela-
tionships and hierarchical control not only at the team level but throughout the
organization. The rise of SMOs has been attributed to more rapidly changing business
environments requiring quicker adaptability, a more professionalized workforce requir-
ing more autonomy, and more advanced IT systems making it possible to coordinate
and manage work without the managers (Bernstein et al. 2016; Lee and Edmondson
2017). The leaders of such organizations tend to believe in empowerment as the way to
get the best and most innovative performance out of employees (e.g., Gelles 2015). The
promise is that this new way of organizing work leads to employee potential being
unleashed as self-directedness and bottom-up innovativeness are able to bloom no lon-
ger constrained by the hierarchical and bureaucratic obstacles. This ought to lead to
significant cost cuts as one no longer needs to have middle management on the payroll
(Hamel and Zanini 2016), while also contributing to a more motivated, innovative, and
loyal workforce (Martela and Kostamo 2017). Buurtzorg not only offers care 20%
cheaper than its key competitors making it financially successful, it also has higher em-
ployee and customer satisfaction than the competitors (Gray et al. 2015). Better per-
formance with lower costs and more engaged employees: Who could resist the
promise?
However, to be able to evaluate whether this really is a new way of designing an
organization, or just another managerial fad, we need to ask what really is novel about
this type of organization. Organizations are “systems of coordinated action”(March
and Simon 1993, p. 2) that somehow need to solve the problems of how the individual
agents are motivated, assigned tasks, and their actions coordinated in order to contrib-
ute to the organizational goals (Puranam et al. 2014). Accordingly, they must find solu-
tions to certain foundational challenges in order to function. Assessing whether a
proposed type of organizational structure is novel is in essence about assessing whether
it has found novel ways to solve one or more of these foundational challenges
(Puranam et al. 2014). Given that such solutions tend to co-occur and reinforce the
utility of each other, thus forming clusters of interconnected structures and practices
(Meyer et al. 1993; Fiss 2007), we are thus asking whether the SMO is a configuration
of novel solutions to these basic challenges.
More particularly, the present article will compare self-managing organizations and
self-organizing (which I use as synonyms) to two established ideal types of
organizational structure, bureaucracy (Weber 1946) and adhocracy (Mintzberg 1980).
In this respect, this article joins a long chain of essays that have aimed to illuminate
organizational life by describing ideal types that then manifest themselves in more or
less pure forms in real-life organizations (Weber 1946; Mintzberg 1980; e.g., Adler
Martela Journal of Organization Design (2019) 8:23 Page 2 of 23

2001). Even though real life is always more complicated than the ideal types (see e.g.,
fig. 1 in Lee and Edmondson 2017), understanding the benefits, drawbacks, and success
factors of these ideal types in various environmental conditions can help practitioners
to make more informed choices about the best ways to design their own organizations.
And even though it might be hard to find a completely “pure”SMO with no traces of
bureaucracy (in the same sense as it is impossible to find a completely “pure”bureau-
cracy), I argue that an SMO as an ideal type is better able to explain the internal con-
figuration of several real-life companies such as Morning Star or Valve than the
existing ideal types that are based on more top-down solutions to the fundamental
problems of organizing. For example, the more than 8000 nurses of Buurtzorg do not
have any supervisors to report to and are supported by fewer than 50 back office people
who take care of salary administration and other administrative tasks (Gray et al. 2015).
Given the complete lack of middle management and manager-subordinate relationships
at Buurtzorg, it is clear that bureaucracy or even adhocracy as organizational structures
are not able to explain how Buurtzorg works. To account for such organizational out-
liers, we thus need a new ideal type: the self-managing organization. And to examine
what really makes it novel, this article investigates what new solutions it has found to
the fundamental problems of organizing, as compared with bureaucracy and adhocracy.
Thus, this research answers not only calls to understand SMOs better (Puranam and
Håkonsson 2015; Bernstein et al. 2016; Lee and Edmondson 2017) but also recent calls
to explore “alternative organizing”and more emancipatory organizational structures
and practices (Grote and Guest 2017; Parker and Parker 2017).
From four to six universal problems of organizing
Following Puranam et al. (2014, p. 163), I see organizations as “(1) a multiagent system
with (2) identifiable boundaries and (3) system-level goals (purpose) toward which (4)
the constituent agent’s efforts are expected to make a contribution.”Each of the four
criteria is taken as critical in the sense that only entities fulfilling all four criteria can be
seen as an organization. Organizations thus are goal-oriented entities constituted of an
identifiable group of independent agents that together contribute towards that goal. To
be able to achieve this coordinated goal-striving, there are four universal problems that
any form of organization must, by definition, solve (Puranam et al. 2014): (1) Task div-
ision: The overarching goal of the organization must be mapped into a set of interre-
lated tasks and subtasks that can be then assigned to individual agents. The division of
tasks can be done consciously in a top-down manner where the larger tasks are divided
into increasingly smaller subtasks or it can be done organically in a more bottom-up
manner where every employee has the authority and responsibility to identify what
tasks need to be accomplished in order to contribute to the overall goal of the
organization. (2) Task allocation: The next step after task division is to map the ob-
tained list of tasks to individual agents and groups of agents. The aim is to ensure that
for each necessary task and subtask, there is some agent or group that is responsible
for ensuring that the task gets accomplished. Again, this allocation can be done in sev-
eral ways using several principles such as specialization (assigning clusters of similar
tasks to the same individuals), streamlining (minimizing interdependence across
agents), or job enrichment (ensuring that the tasks are diverse enough and lead to
Martela Journal of Organization Design (2019) 8:23 Page 3 of 23

tangible output in order to be interesting and motivating, see Hackman and Oldham
1980).
While the two first universal problems are about the division of labor, the latter two
are about the integration of effort that needs to be accomplished when individual
agents start fulfilling their tasks. (3) The provision of rewards: Given that individual
agents have their own interests and values, the organization needs to provide some
form of reward to make the agents motivated to accomplish their tasks. The most
prototypical form of reward in organizations is monetary compensation, but the agents
can also be motivated by many other factors such as status and advancement opportun-
ities, or more intrinsic motivations such as self-expression and contributing to a worthy
goal through one’s work (Deci and Ryan 2000; Martela and Kostamo 2017). One way
or another, the organization must ensure that the constituent agents have the necessary
motivation to accomplish their tasks and contribute to the organizational goal. (4) The
provision of information: To successfully accomplish their tasks in a manner that ad-
vances the overarching goal, the individual agents need constantly updated information
about the environment and about what the other organizational agents are doing. In
particular, several tasks are often interconnected and the agents need to coordinate
their actions in order to be able to accomplish them.
Puranam et al. (2014, p. 166) argue that any form of organization “must necessarily
feature solutions to each of the four basic problems of organizing.”While an
organization has to deal with other problems, especially adaptation to environment, in
order to endure and be successful, without solving these four problems it does not even
get in the game where it’s potential for adaptation is tested. Finding solutions to these
four problems transforms a group of individuals into an organization, and in this sense
are, according to Puranam et al. (2014, p. 166), “sufficient for an organization to exist”:
“If the four problems have proposed solutions, then we have in effect (a) a set of
tasks believed to contribute toward a goal (b) assigned to a set of agents who have
been (c) rewarded for and (d) informed about executing those tasks so that their
efforts are expected to contribute toward the goal of the organization.”(Puranam
et al. 2014, p. 166)
In comparing SMOs with bureaucracy and adhocracy, I will base the examination on
these four universal problems of organizing. However, to add more granularity to the
comparison, I will divide two of these four problems both into two sub-problems. This
is done because I believe the sub-problems will be illustrative in highlighting what
makes SMOs special as compared with other organizational structures. First, provision
of rewards aims to solve the problem of how to “motivate the agents to cooperate”
(Puranam et al. 2014, p. 165). For this, one firstly needs to have some inducements or
rewards for the agents that motivate them to act in ways that fulfill the organizational
goals. The other side of the coin is that there must also be some mechanisms that en-
sure that these rewards are given only to those deserving them. Freeriding is a universal
problem in any collective human action (Simon 1991; Fehr et al. 2002). Given the pri-
vate interests of individual agents, there is always a risk that someone is free-riding by
reaping the rewards without making the contribution—or outrightly using the collective
resources for their own personal benefit. In addition to corruption or stealing, someone
Martela Journal of Organization Design (2019) 8:23 Page 4 of 23

might exploit the system by getting the same compensation without putting in the
same effort as others. In economic public good games, it has been shown that without
any option to punish free-riders, cooperation quickly declines (Fehr and Gächter 2002).
When people see that others ‘get away with it’, they start to question why they should
put in the effort. Similarly, if an organization is unable to address self-serving and free-
riding, morale can quickly deteriorate leading to an escalation of self-serving behavior.
Thus, in order to function, any organization must have both an incentive structure re-
warding desired behavior and effective means of addressing freeriding. Both are essen-
tial parts of efficient provision of rewards, and as we will see, SMOs offer interesting
solutions to both of these sub-problems.
Second, Puranam et al.’s(2014)provision of information problem will be divided into
two sub-problems: providing direction and ensuring coordination. First, when individual
employees execute their work, it must be ensured that what they do and the decisions
they make contribute to the overall goals of the organization. In other words, they need
some direction in regard to what actions and decisions are the right ones. Without such
direction, there is a very real danger of sub-optimization where people make decisions
that make sense from their partial point of view, but that are actually disadvantageous
on the organizational level. Thus, the organization needs to ensure that each individual
has some guidance mechanisms that ensure that their actions and decisions are benefi-
cial from the point of view of the whole organization and its goals. Second, given that
the successful accomplishment of many singular tasks by one agent is often inter-
dependent on the progress of other tasks carried on by other agents, there need to be
mechanisms that ensure the coordination of these interdependent tasks. Although both
of these sub-problems are about the provision of information, how to ensure the coord-
ination of interdependent tasks and how to ensure that people’s actions and decisions
contribute to the organizational goals are relatively independent problems, requiring
different kinds of solutions. Thus, in comparing SMOs, adhocracies, and bureaucracies,
it will be informative to examine separately how these three types have solved both of
these sub-problems.
Accordingly, we arrive at a list of four main problems, two of which are divided into
two sub-problems that any organization needs to solve in order to function (see Table 1),
and that will be used as a base for comparing bureaucracy, adhocracy, and self-managing
organizations.
Comparing bureaucracy, adhocracy, and self-managing organizations as
solutions to the universal problems
Bureaucracy as an organizing paradigm (see Weber 1968 especially chapter XI)
emerged as a replacement to the older forms of organizing, which had been based on
“individual privileges and bestowals of favor”(p. 958), where a person “owns”one’s pos-
ition and can take advantage of it as one wants, thus blending personal and
organizational property and loyalties. Instead of such organizing based on personal rela-
tionships, loyalties and blood ties, bureaucracy is based on impersonality: Individuals
occupy certain roles and their responsibilities and rights are determined by strict and
explicit rules. As classically outlined by (Weber 1968, pp. 956–958; see also Hall 1963),
such bureaucracy is based on six principles: (1) There are fixed official jurisdictional
areas ordered by rules, (2) there is a well-defined hierarchical system of authority, (3)
Martela Journal of Organization Design (2019) 8:23 Page 5 of 23

administration is based on written documents, (4) management presupposes expert
training, (5) taking care of one’s office requires the full working capacity of the official,
and (6) the management of the bureaucracy “follows general rules, which are more or
less stable, more or less exhaustive, and which can be learned”(p. 958, emphasis in ori-
ginal). The last point emphasizes the embedded idea of rationality within bureaucracy:
The manager—given the right training—can control, govern, and lead the whole
organization in a rational manner. The apparent complexity of the organization and its
relations to its environment can be reduced to such a degree that a single person (or a
Table 1 How the three organizational forms solve the six fundamental problems of organizing
Weberian bureaucracy Mintzberg’s adhocracy Self-managing
organizations
Basic structural
arrangement
Hierarchical Matrix Flat
Manager
power over
subordinates
Extensive Loosened, especially
within-teams
(Almost) non-existent
Decision-making
power
Hierarchical Selective decentralization Radical decentralization
Division of
labor: Task
division
Top-management identifies
the necessary tasks, which
are then divided into sub-
tasks and sub-sub-tasks by
layers of management in a
top-down process.
Top-down process. But
within-team task division is
done more autonomously,
with managers retaining
the right to intervene and
authorize larger decisions.
Bottom-up and top-down
with top management as
well as individual
employees having the
responsibility and authority
to create new tasks.
Division of
labor: Task
allocation
Top-down process where
tasks and responsibilities are
allocated by managers to
their subordinates.
Within-team allocation done
autonomously, but allocation
of employees to teams and
other major resource
decisions done by a
top-down process.
Emerges from the
interaction between
employees who have
considerable authority to
decide which roles to fulfill.
Provision of
reward:
Rewarding
desired behavior
Monetary compensation,
bonus systems, and
advancement options
allocated by supervisors.
Monetary compensation with
sometimes team-based bonus
systems allocated by supervisors.
The chance to work in more
interesting projects in the
future.
Monetary compensation
with salary levels and
bonuses often determined
through a peer-based
process. Much emphasis on
making the work
intrinsically rewarding.
Provision
of reward:
Eliminating
freeriding
Supervisors responsible for
monitoring the actions and/
or outputs of their
subordinates.
Supervisors responsible for
monitoring the actions and/or
outputs of their subordinates.
Within teams employees might
be primarily accountable for
each other.
Employees primarily
accountable for and
monitored by each other
with explicit conflict
resolution mechanisms to
resolve conflicts.
Provision of
information:
Direction setting
Strict task boundaries and
precise instructions eliminate
the need for distributing
broad information.
Teams typically have necessary
information to make most task-
related decisions. Other
decisions made at higher
levels where the managers
have enough overall
information for
decision-making.
Transparency of all key
information and active
distribution of information
to ensure each employee's
capability to make decisions
that benefit the wholeness.
Provision of
information:
Coordination of
interdependent
tasks
Standardized procedures
and top-down task
allocation and monitoring
ensure the necessary
coordination.
Within-team coordination
typically done through
constant communication.
Within-function coordination
accomplished hierarchically,
with interfunctional teams
helping the functions to
coordinate.
Within-team coordination
accomplished through
constant communication.
Within-team and within-
function coordination ac
complished through coord
ination roles and through
extensive use of IT systems.
Martela Journal of Organization Design (2019) 8:23 Page 6 of 23

small group of top managers) with “a special technical expertise”(p. 958) can identify
the law-like general rules through which the organization can be administered in the
most efficient and rational manner. Equipped with this rationality and expertise they
then divide the necessary tasks to their subordinates, who in turn divide them to their
subordinates, and thus, the will of the top-level experts is executed in the hierarchical
organization, where those at the lower levels are entrusted with highly specialized and
fixed tasks as part of a mechanism that “cannot be put into motion or arrested by him”
(Weber 1968, p. 988). All that the subordinates need to know and care about is fulfilling
the duties and obligations of their specific roles, with the boundaries and goals of these
roles determined by their superiors. Classically it is seen that the resulting alienation that
the workers often feel is the price one has to pay for efficiency (see Adler 2012).
While bureaucracy in its most classic form was found in the factories of the early
20
th
century, Mintzberg argued that after World War Two adhocracy, a new ideal type
of organizational structure, emerged as it was better able to answer the call for more in-
novativeness in dynamic environments (Mintzberg 1980). Almost forty years ago he ar-
gued—citing examples such as The Boeing Company and NASA—that “adhocracy
seems clearly to be the structure of our age”(Mintzberg 1980, p. 338 emphasis in ori-
ginal). Mintzberg defined adhocracy as an organizational structure where highly trained
experts work in multidisciplinary teams producing unique outputs, and where the
organization is relatively decentralized and coordination is achieved largely through
mutual adjustment (Mintzberg 1980; Mintzberg and McHugh 1985). These firms tend
to operate in environments that are both dynamic and complex, thus requiring fairly
sophisticated levels of innovation in their outputs. Typical examples are consulting
firms, advertising agencies, film companies, and space agencies. Adhocracy thus seems
to take several steps away from classic bureaucracy towards self-organizing. However,
at the same time, Mintzberg emphasizes that adhocracies still rely “extensively on
matrix structure”(Mintzberg 1980, p. 337), where traditional line organization exists in
tandem with project-based teams. Although “quasi-formal authority”is given to staff
personnel, thereby “blurring the line-staff distinction”(Mintzberg 1980, p. 337), the
supervisor-subordinate relationship is not eliminated and decision-making authority is
not fully given to the personnel, thus making this form of organizational structure
clearly distinct from that of an SMO. Galbraith’s study of Boeing (Galbraith 1973),
which Mintzberg (1980) cites as one of the prime examples of adhocracy, describes
how the company adopted a functional form with some cross-functional teams and task
forces enabling the necessary coordination, but within each function the design of the
aircraft was placed “under a single authority structure”(p. 123) with hierarchical chains
of command and functional managers making decisions “relating to who would do the
work and how to do it”(p. 133). The plastics organizations in Lawrence and Lorsch
(1967), another key example of adhocracy according to Mintzberg (1980), had a very
similar functional form with dedicated teams handling integration between functional
units. Adhocracy thus introduces matrix-like elements into the organization with cross-
functional teams having the authority in certain decisions and with teams given more
independent decision-making power within certain boundaries. Instead of one clear
chain of command, there are a few competing chains of command, but the essentially
hierarchical manager-subordinate authority relationships are still retained, with man-
agers having the right to intervene in most decisions made by their subordinates.
Martela Journal of Organization Design (2019) 8:23 Page 7 of 23

If adhocracy was in vogue in the 1980s, one could claim that the self-managing
organization is the organizational structure that has currently captured the headlines as
the next big thing in designing organizations (e.g., Hamel 2014; Gelles 2015). Following
Lee and Edmondson (2017), SMOs are defined by three characteristics: (1)
Decentralization of authority is radical rather than incremental. Instead of delegating
some authority to employees, in SMOs the hierarchical reporting relationship between
the manager and the subordinate is (almost) completely eliminated and the employees
have full authority to make key decisions as regards their work. For example, at Valve,
employees can even “ship”their own products provided that two or more other em-
ployees agree (Puranam and Håkonsson 2015). The decentralization also applies to re-
source control, a key dimension of power within organizations (Astley and Sachdeva
1984). The employees of SMOs typically can make independent or team-level decisions
on the purchase of resources that would require permission from above in more bur-
eaucratic organizations. 2) The decentralization of authority takes place throughout the
organization, with only a handful of key decisions left to the top management. Instead
of pockets of self-management within hierarchical structures—such as self-managing
teams typical in adhocracies—the decentralization of authority and the elimination of
hierarchical relationships is effective throughout the organization. (3) The
decentralization of authority is formal and systematic. Informally delegated authority
can be reversed at any point (see Foss 2003), the threat of which can easily constrain
subordinates even when the reversal of authority is not actualized. In SMOs, the
decentralization of authority is codified in explicitly articulated organizational principles
that institutionalize the self-managing way of working, in effect prohibiting the man-
agers from exercising certain forms of authority. SMOs thus are organizations that have
formally and systematically decentralized authority throughout the organization to the
degree of having abolished supervisor-subordinate relationships and layers of middle
management almost completely (Lee and Edmondson 2017).
Having identified the key characteristics of bureaucracies, adhocracies, and SMOs, it
is time to investigate how they solve the six fundamental problems of organizing en-
listed above (see Table 1).
Task division
In classical bureaucracy, top-level management provides the task architecture. Having
established the organizational goals, their expertise allows them to identify the neces-
sary tasks needed to accomplish these goals. In order to be able to control the whole
process, they thus divide the whole operation into rather fixed official jurisdictional
areas with detailed specifications about what each area needs to accomplish and how.
The managers of each of those areas then further divide the task into smaller sub-tasks,
making the task division into a top-down process.
In adhocracy, the higher-level task architecture is also provided by the top manage-
ment but, within certain boundaries, teams can have relatively unrestricted autonomy
to establish the sub-goals and tasks needed to achieve the overarching goal of the team.
Sometimes they can also set precedents that can affect the direction of the whole
organization. For example, the consulting team working with a particular client might
note a “unique challenge requiring a creative solution”(Mintzberg 1980, p. 337), and in
Martela Journal of Organization Design (2019) 8:23 Page 8 of 23

adhocracy, it has the authority to decide how best to address that challenge. However,
the authority given to the teams is “quasi-formal”and the decentralization is “selective”
(Mintzberg 1980, p. 337), meaning that the freedom of the teams has clear boundaries,
beyond which they have to seek permission from their supervisors, who also retain the
right to intervene at any moment (cf. Foss 2003).
In SMOs, given that distributing decision-making power is their defining feature, it
follows that task division is also more distributed. Although the top management might
provide the overarching structure, individual employees and teams have the responsi-
bility to identify new tasks that need to be accomplished to serve the organizational
goals, and the authority to make these tasks part of the organizational structure. At
Valve, “self-selected teams of individuals form spontaneously around topics of interest”
(Puranam and Håkonsson 2015, p. 3) and temporary role structures emerge organically
within each team, and at Zappos, “a formal governance process enables anybody, even
the most junior members, to propose changes to the roles, accountabilities, policies and
decision rights in their work group”(Lee and Edmondson 2017, p. 47). At Morning
Star, there are no centrally defined roles, and the employees are even made responsible
for initiating the hiring process when they find themselves overloaded or spot a new
role that needs filling (Hamel 2011). This way a significant part of the task division
emerges organically from the bottom-up rather than being imposed top-down.
One implication of this is that employees at SMOs need to be both willing and cap-
able of taking much broader responsibility for business success than in traditional orga-
nizations. Instead of just executing the tasks allocated to them from above, the
employees need to be on the lookout for what new tasks need to be accomplished for
better results. This requires an understanding of the broader goal of the organization in
order to be open to finding out what new things could advance those broader goals.
The proactivity of employees is thus crucial for successful task division at SMOs, and
as work engagement contributes to proactivity and personal initiative (Salanova and
Schaufeli 2008; Hakanen et al. 2008), this puts pressure on these companies to ensure
that each employee is highly engaged and motivated. In organizations where task struc-
ture is given to an employee from higher up, that employee can successfully execute
tasks without having to be particularly proactive, but in SMOs, where the responsibility
to identify necessary tasks is to a large degree on employees, higher levels of proactivity
become almost a necessity.
Task allocation
In a bureaucracy, task allocation is also done in a top-down manner. Having established
the necessary broader tasks for the accomplishment of organizational goals, the top
management allocates these to the next level of management, which in turn makes the
allocation decisions in regard to their subordinates. Thus, each manager is responsible
for making the task allocation for the layer of employees directly below themselves.
This task allocation is typically done in a meritocratic manner, aiming to match tasks
and sub-tasks to employees’skill profiles.
In regard to adhocracy, the overarching task allocation is also done in a top-down
manner. However, teams typically have the authority to divide the tasks among them-
selves as they wish, thus creating pockets of more distributed decision-making power
Martela Journal of Organization Design (2019) 8:23 Page 9 of 23

within the more hierarchical allocation process. But in major decisions such as recruit-
ing new members to accomplish tasks that the current members do not have the time
or skills to accomplish, they typically need to get approval from higher up in the hier-
archy, and the decision about which employees are allocated to which teams are typic-
ally also made by the managers above the teams.
SMOs are defined by the decentralization of decision-making power, and this
decentralization also applies to task allocation, one of the key dimensions of decision-
making power. Thus, for several types of tasks, one would expect that SMOs have given
to individuals or teams the authority to choose what tasks are done by whom. For ex-
ample, Valve gives employees the ability to choose which projects they want to work
on (Puranam and Håkonsson 2015). And Buurtzorg allows each team to decide within
the team how various tasks are allocated. This requires that the employees are commit-
ted enough to the organizational goals to want to make sure that even the less attract-
ive, but necessary, tasks get accomplished.
Such distributed allocation of tasks is probably easier to implement when there is not
too much interdependence between teams/units. If team X works with customer A and
team Y works with customer B, and their work is mostly not dependent on what the
other team is doing, it is relatively easy to give both teams the authority and independ-
ence to allocate tasks within their customer project as they wish. Both Valve and Buurt-
zorg work in industries where each team works quite independently, not being
dependent on the input of other teams.
If, however, teams X and Y work in a factory where team Y’s capability to exe-
cute their own work is significantly dependent on team X accomplishing certain
tasks in a certain order, then distributed task allocation is not so straightforward.
One needs to ensure that somebody actually executes the tasks that one’sown
work is dependent on. And in the absence of supervisors ensuring this, one is re-
quired to invent other mechanisms. For example, at the heart of Morning Star’s
way of organizing work at their tomato processing factories are Colleague Letters
of Understanding (CLOU) that each employee negotiates with the associates who
are most affected by one’swork(Hamel2011). The CLOUs list the responsibilities
(with relevant performance metrics) that the two employees have towards each
other, and can cover as many as 30 activities. Most employees have about 10 col-
leagues with whom they negotiate such CLOUs. The task allocation is built around
the several thousand formal relationships between the employees that are formed
this way and that are renegotiated regularly. This ensures that each necessary task
gets allocated to somebody. But instead of a top-down process, the employees thus
allocate each other to certain tasks and roles through mutual negotiations.
Provision of rewards: rewarding desired behavior
In a bureaucracy, there is typically no assumption that the tasks would be intrinsic-
ally rewarding, given that alienation from one’s work is seen as one of the unfortu-
nate side effects of increased rationalization and efficiency (see Adler 2012).
Accordingly, monetary compensation in the form of salaries and bonuses is the key
reward mechanism. And decisions about compensation are made in a top-down
manner, with supervisors deciding who gets a raise or who gets promoted to a
Martela Journal of Organization Design (2019) 8:23 Page 10 of 23

higher position in the organization. Given the hierarchical structure, many em-
ployees find motivation in a chance to climb the corporate ladder—to have “a‘car-
eer’within the hierarchical order”(Weber 1968, p. 963). To achieve more power,
autonomy, status, and compensation in the future the workers aim to offer a dili-
gent performance in the present day.
Also in adhocracies, monetary compensation is the key reward mechanism, with su-
pervisors deciding compensation levels and allocating bonuses. Given that work is often
done in teams, bonus systems are sometimes team-based rather than individual-based.
However, given the matrix structure and often project-based work where many experts
occupy specialist roles, there might not be clear pathways for promotion. Instead, the
employees might be motivated by the ability to work in more interesting projects and
having more responsibility in the future.
In regard to SMOs, given that they often lack the supervisors that can make salary
decisions and other rewarding decisions, a different solution to this problem must be
found. The key is to identify which body in the organization would have enough infor-
mation about individual employees to make accurate decisions about the salary levels
they deserve as well as other rewards. In some industries, it might be possible to have
quantitative performance metrics that are so accurate that one could base the reward
mechanisms entirely on them. However, in many industries, especially those requiring
expertise, qualitative results, and collaboration, the quantitative metrics might not cap-
ture a large part of the employees’actual work and contribution to the organization. In
these situations, the most reliable source of information about an individual employee’s
performance is often her or his coworkers. Accordingly, it is expected that various
peer-based compensation and reward mechanisms would be the most typical way to
solve the problem of rewarding employees in SMOs. For example, at Morning Star
compensation is determined through a process of peer feedback with ultimate decisions
made by an elected local compensation committee whose members are elected by col-
leagues (Hamel 2011). Similarly, Valve has a peer-reviewed performance system where
employees rank each other’s performance and top performers receive generous bonuses
and raises (Puranam and Håkonsson 2015). Given the lack of managers in a position to
evaluate the employees, the compensation decisions thus have been allocated to those
people who directly see and are affected by the employee’s performance, i.e., one’s
colleagues.
Furthermore, SMOs typically rely more on intrinsic motivation than classical bur-
eaucratic organizations (Martela and Kostamo 2017). They are built on the as-
sumption that people are not passive and lazy, only moved to action by extrinsic
rewards, but instead active and willing to perform well if only given a chance and
a task that they find intrinsically motivating (McGregor 1960; see Deci et al. 2017).
Thus, Buurtzorg is built on the assumption that nurses want to serve their patients
to the best of their ability and should be given the freedom to do so. This mission
has served them well as they have won the “Best Employer”award in the
Netherlands four times in the 2010s (Beste Werkegevers 2017). At Morning Star,
each employee is responsible for drawing up a personal mission statement that
aligns one’s own mission with that of the company (Hamel 2011), while Zappos is
dedicated to creating an especially fun place to work (Gelles 2015). Giving em-
ployees significant autonomy and freedom tends to require that motivation of the
Martela Journal of Organization Design (2019) 8:23 Page 11 of 23

employees is not only based on extrinsic rewards and punishments but on an in-
trinsic willingness to perform one’s tasks well (Martela and Kostamo 2017).
Provision of rewards: eliminating freeriding
In a bureaucratic organization, a strict hierarchical structure aims to ensure em-
ployee compliance through supervisors monitoring their subordinates’output and
actions. The supervisor provides the goals and tasks, and a time frame of when
they need to be accomplished, and the subordinates are held accountable for being
able to deliver the necessary work in time. In addition to holding employees ac-
countable for completing the tasks provided by the supervisor, bureaucratic organi-
zations often include explicit ways to monitor the whereabouts of the employees
such as a time clock to record their presence in the workplace and the supervisor
visually observing that they are actually doing work. The ideal is to be able to
eliminate through monitoring the possibility that the employees could somehow
evade doing the work they are paid to do. This might sound rational and effective,
but it is important to remember that the more the employees are observed, the
more they tend to hide things (Bernstein 2012,2017).
In adhocracy, the employees are also held accountable for their results by their super-
visors, and similar monitoring techniques are used. The monitoring becomes slightly
more complex as in a matrix organization an employee might be held accountable by
two separate managers, requiring more coordination regarding what amount of time
one is supposed to spend serving the separate tasks administrated by separate man-
agers. Furthermore, while teams are accountable for their managers, within individual
teams, the team members might be primarily accountable for each other to accomplish
the assigned tasks, and thus, peer control might replace supervisor control on a within-
team level (cf. Barker 1993).
In SMOs, the lack of supervisors again means that monitoring must be conducted by
those parties who are best able to see whether the employee is doing what is expected.
Firstly, this means that the emphasis is typically not in monitoring the activities of the
employees—they are often free to choose when and how they work—but in monitoring
their performance using the available quantitative metrics. Furthermore, for many as-
pects of performance that are not captured by these metrics, it is typically one’s peers
who are in the best position to evaluate whether the employee is fulfilling one’s respon-
sibilities. Thus, the potential lack of performance or effort is expected to be monitored
by one’s colleagues. For example, Valve uses a system of peer-driven employee
evaluations (Lee and Edmondson 2017), and at Morning Star, through a system of
mutual contracts, each employee becomes “a contractor in a web of multilateral
commitments”, with each employee expected to hold one another accountable for
results (Hamel 2011, p. 52).
One implication of this is that individual employees need the skills to handle situa-
tions where one party fails to fulfill one’s responsibilities. As they cannot ask the super-
visor to intervene, they need to be able to intervene themselves. Laloux (2014, p. 114),
in investigating several organizations that manifested principles of self-organizing,
found that many of them had converged around “virtually identical conflict resolution
mechanisms”that typically involved a few formalized steps that the employees have to
Martela Journal of Organization Design (2019) 8:23 Page 12 of 23

take with each other when they find themselves in conflict over fulfillment of responsi-
bilities or other issues. The spirit of SMOs in ensuring that the work gets done is cap-
tured by one Morning Star employee that Hamel (2011, p. 53) quotes: “Around here,
nobody’s your boss and everybody’s your boss.”
Provision of information: direction setting
In a bureaucratic organization, key information is concentrated at the top, and the task
division aims to be so specific that individual agents do not have to be aware of the big-
ger picture in order to fulfill their tasks. They are “a small cog in a ceaselessly moving
mechanism which prescribes to him an essentially fixed route of march”(Weber 1968,
p. 988). The traditional assembly-line worker does not have to know anything about
the organizational strategy because top-down task allocation and instructions ensure
that the tasks they execute contribute to the organizational goals. As long as they ac-
complish the specific tasks and roles assigned to them from the above, the employees
will contribute to the organizational goals, whether they are aware of those goals or
not.
In adhocracies, the teams typically face more unique and complex problems and
often have among themselves the best expertise to make decisions about how to solve
these problems. Thus consulting teams working with a customer typically get the ne-
cessary information to make decisions straight from the customer. However, the teams
typically concentrate on the task at hand, without having to think about the overall
strategy of the organization and thus do not typically require information about this ei-
ther. At the higher levels of the organization, adhocracies typically rely on the same
mechanisms as bureaucracies: Subordinates report to their supervisors who report to
their own supervisors, and key decisions are done at higher levels where the managers
have enough overall information to make such decisions. Furthermore, the supervisors
who are in a better position to see the wholeness typically “reviewed decisions made by
their subordinates to make certain that specialists working on one part of the product
line were not doing anything that would adversely affect another part”(Lawrence and
Lorsch 1967, p. 144).
In SMOs individual employees are given authority to make independent decisions on
matters and purchases that would in a bureaucratic or adhocratic organization need to
be approved by someone higher up in the managerial chain. As Hamel (2011, p. 53) de-
scribes, “if a maintenance engineer needs and $8,000 welder, he orders one.”The bigger
the decision the individual employee is able to make, the wider are its implications for
the organizational wholeness. Accordingly, in order to make those decisions in a way
that is beneficial for the organizational wholeness, the employees must have enough in-
formation about the organizational wholeness. Thus, in empowering employees to
make such decisions SMOs need to ensure that each employee has enough information
about the whole situation of the organization and its environment in order to make
wise decisions from the point of view of this wholeness. As Martela and Kostamo
(2017) argue, in bureaucratic organizations typically only top management has the cap-
acity to make decisions that benefit the wholeness, because only they have up-to-date
information about, for example, the current financial situation of the organization.
When decision-making power is decentralized, the information distribution must be
Martela Journal of Organization Design (2019) 8:23 Page 13 of 23

decentralized as well, to avoid employees making suboptimal decisions due to lack of
relevant information. Thus, a wide distribution of information is necessary for self-
organizing to succeed.
Accordingly, to ensure that individual employees can make decisions that are benefit-
ing the wholeness, self-organizing organizations typically make all information, includ-
ing financial information, transparent to ensure that every employee has clarity over
the current organizational situation and goals. Transparency here is thus about disclos-
ure—making previously restricted information available for employees (Bernstein
2017). For example, at Morning Star each business unit publishes detailed financial ac-
counts twice a month that are available to every employee (Hamel 2011). Wider access
to information also means that the employees need to have the necessary skills to use
and understand that information in their decision-making. For example, if they need to
take the financial situation of the company into account in some decisions, they need
to be able to analyze the financial information to determine the situation. This puts in-
creased pressure on SMOs, compared with traditional organizations, to ensure either
through recruitment or through internal training that each individual has the necessary
skills to analyze and understand the information that they need for their decision-
making. This is one reason why SMOs can be expected to have more educated em-
ployees than traditional organizations. All in all, in SMOs, transparent and effective dis-
tribution of information is a key enabling condition to ensure autonomous-yet-wise
decision-making.
Provision of information: ensuring coordination
To ensure necessary coordination, bureaucratic organizations tend to implement a set
of prescribed procedures using formal communication. Directives, schedules, plans, and
standards aim to eliminate—as much as possible—any need for spontaneous, ad hoc
coordination (Puranam et al. 2014, p. 166). Coordination is thus ‘rationalized’by man-
agement expertise that designs the procedures and steps that individual employees
must follow in order to establish necessary coordination. An individual employee can
trust that if one just follows the procedures and standards, one’s actions will align with
those of others.
In adhocracy, within-team coordination is typically established with the team spend-
ing face-to-face time with each other, thus ensuring that everybody is ‘on the same
page’and knows what others are doing. For example, at Boeing, the units working on
the same section of the aircraft were physically located together (Galbraith 1973). In
operating adhocracies (Mintzberg 1980, p. 337), such as consulting firms, IT firms, ad-
vertising agencies, or film companies, there typically is not much need for cross-team
coordination as each team works with a separate project. In administrative adhocracies,
such as chemical firms or space agencies (Mintzberg 1980, p. 337), the administrative
component functions as an adhocracy, but the operating core below it works in a more
bureaucratic manner. For example, at Boeing an interfunctional team chaired by the
program manager was used to coordinate when each function had to deliver certain
key outcomes. In addition, various interfunctional task forces and teams were used to
ensure the lower-level coordination needs between the functions. But within each func-
tion “rules, hierarchy, goal setting and planning, and direct contact”(Galbraith 1973,p.
Martela Journal of Organization Design (2019) 8:23 Page 14 of 23

134) were still used in a more bureaucratic manner to ensure within-function
coordination.
In SMOs as in adhocracies, individual teams achieve coordination through close
proximity, regular meetings, and regular electronic communication. The key difference
is in how the coordination between teams is organized. While adhocracies rely on a
combination of hierarchical structures and interfunctional task forces, SMOs typically
lack the possibility to rely on the former and accordingly need to rely on the latter.
When the coordination between two teams or units requires constant interaction, spe-
cific individuals and groups are given the responsibility to act as links between them.
Here necessary coordination is thus achieved through specific individuals engaging in
enough communication between two groups to be able to act as intermediaries.
Furthermore, another option open for SMOs to achieve coordination is to use various
IT systems, software, or even physical artifacts like Kanban boards for this task. The
rise of such programs and systems, in fact, might be one of the contributing factors to
the rise of SMOs in the first place, as they have been able to take over many key coord-
inating needs of organizations traditionally performed by managers. For example, the
coordination of work at Zappos relies to a significant degree on “Glass Frog,”a dedi-
cated software meant to assist in “adoption, record keeping, and ongoing practice”of
self-organizing (Robertson 2016, p. 154). Glass Frog and similar systems at other SMOs
track the progress of various individuals, teams and projects, and keep tabs on who is
responsible for what, and through being accessible to all employees, make sure that
each employee can check the situation of any project or process that they need to be
aware of for their own task fulfillment. Thus these systems take over much of the co-
ordination, progress monitoring, and information distribution tasks that traditionally
managers have taken care of.
Boundary conditions: in what contexts is self-management likely to succeed?
Self-managing organizations thus seem to offer a different way to structure an
organization compared with the more traditional alternatives. But what we need to ask
is in what industries and under what conditions is it a more viable option compared
with other organizational structures. In other words, what are the boundary conditions
under which self-managing organizations could outperform bureaucracies or adhocra-
cies? To answer this question, I will identify three factors that are likely to influence
how easy it is to implement self-management and how successful it might be when
implemented.
Independence vs. interdependence between units
The relative independence of units from each other affects the coordination challenge
that an SMO faces. If there is very little need for mutual coordination between units or
teams, it is relatively easy to give teams independence to make their own decisions and
be responsible for their own performance. For example, at Buurtzorg, nurses working
within a team taking care of one region of the Netherlands need to do very little coord-
ination with other teams working in other regions. Each team serves its own customers,
without having to know what other teams are doing—and thus, it is possible to give
them almost complete autonomy to make decisions in their own region. In contrast, in
Martela Journal of Organization Design (2019) 8:23 Page 15 of 23

a factory producing airplanes, one unit’s decisions will significantly affect other units
downstream from it. The ability of a certain unit to accomplish its tasks is dependent
on getting the necessary input such as parts and bolts from other units, requiring the
units to engage in a high level of mutual coordination, information sharing, and sched-
uling, which is accomplished easiest through top-down oversight. This is not to say that
it would be impossible to implement self-management even in contexts of high unit
interdependence—the factories of Morning Star seems to be an example of this—just
that in the latter case the challenges of coordination that need to be solved are much
more complex, thus leading most organizations to default to tried and tested coordin-
ation mechanisms offered by bureaucratic organization.
Output standardization vs. customization
If the products or the services of the organization are highly standardized, it is easier to
apply the bureaucratic model of task separation and process optimization, leading to in-
dividual employees being given clearly defined tasks that they can execute without hav-
ing to know much about the context or things outside their strictly defined role. If,
however, each customer is served a highly tailored and unique product, such top-down
standardization easily produces obstacles that make it harder for the teams to serve the
customers well. In these situations, it makes sense to decentralize decision-making
power to allow each team serving their own customers enough authority to make the
necessary decisions that ensure that they can offer the customers the best possible
products and services. In other words, in markets where customers demand highly cus-
tomized and unique products, organizations with empowered teams that have the
means to respond to customer demands rapidly and accurately are likely to outcompete
teams whose ability to serve the customer right is more restricted by company policies
and having to ask and wait for approval to any deviations from the standard.
Employee expertise and motivation
SMOs, in minimizing the power of supervisors to control employees, depend on em-
ployees to be both willing and able to put in the effort for organizational goals. When
there is less monitoring of employees, one is more dependent on the self-motivation of
the employees. And when there is no supervisor to guide and command the employees,
one needs to rely on the expertise of the employees to be able to make the right
choices. Thus SMOs are likely to function better in sectors where employee expertise,
motivation, and work ethic are high. In fact, when proactivity is high, too much super-
visor oversight and control could prove to be demotivating (Grant et al. 2011), making
SMOs especially attractive for industries with a proactive workforce. Traditional strong
professions such as medical doctors, nurses or lawyers, where entry into the workforce
requires an advanced educational degree, could thus be places where one could expect
SMOs to succeed. On the other hand, in low-salary, low-skill industries where most
employees are only motivated by the paycheck and have no loyalty towards the profes-
sion, implementing self-management is prone to lead to lower performance and freerid-
ing. In such industries, the constant monitoring of employees might be necessary to get
adequate output out of the employees. The organization also needs to ensure that em-
ployee motivation is directed towards serving the interests of the organization. While
Martela Journal of Organization Design (2019) 8:23 Page 16 of 23

an expert might be highly motivated to perform certain specific tasks well, self-
managing requires these employees to also look at the overall picture to ensure that
they are performing tasks that benefit the overall goal of the organization. Thus, in
addition to loyalty to their profession, the employees need to have adequate amount of
loyalty towards the organization. This means that having a mission worth committing
to and being able to communicate it to employees might be especially important to the
success of self-managing organizations. Moreover, given the need to ensure that em-
ployees are highly motivated and skilled, SMOs most likely need to offer wages above
the average wage level in their industry to ensure that they attract the motivated and
talented workforce that they require. For example, it is reported that Buurtzorg pays its
nurses above industry average.
All in all then, I argue that we are more likely to see successful SMOs in industries
where teams can work relatively independent from each other, where output is tailored
to each customer, and where employees are highly motivated and skilled professionals,
compared with industries where interdependencies between units are high, where out-
put is standardized, and employee motivation and skill levels are low. It is worth noting
that these obstacles might not be insurmountable, but these three boundary conditions
can help to explain the most likely contexts where SMOs could have a competitive ad-
vantage over bureaucracies, and in what contexts a more bureaucratic structure could
still provide an advantage. For example, Valve produces video games, which is an in-
dustry where each product is unique and highly tailored, where one team working on
one game does not have to do much coordination with other teams, and where em-
ployees typically are highly motivated and the competition for the best talent is fierce.
Similarly, Buurtzorg operates in an industry where each customer requires individually
tailored care, where teams of nurses serving different neighborhoods do not need to co-
ordinate their efforts with other teams, and where employees are part of a strong pro-
fession with certified expertise and high work ethics. Despite the apparent differences
between video game producing and home care delivering, they thus both represent in-
dustries where SMOs are most likely to succeed.
Discussion
This article has aimed to compare how Weberian bureaucracies, Minztbergian adhocra-
cies, and self-managing organizations answer four fundamental challenges every organi-
zations needs to solve, two of which I divided into two sub-challenges. As this
investigation shows, the SMO model seems to offer novel ways of solving these prob-
lems and thus can be considered a new form of organizational design as compared with
the more established alternatives (Puranam et al. 2014). As compared with bureaucratic
organizational design, adhocracy loosens some bureaucratic constraints and leads to
some decentralization of authority, especially at the team level. At the same time, as
compared with SMOs, adhocracy retains some of the hierarchy, controlling, and chains
of command associated with bureaucracy. Adhocracy, as outlined by Mintzberg, thus
seems to occupy a space somewhere in between classical bureaucracy and fully-
developed self-organizing. This emphasizes the fact that we should treat both bureau-
cracy and self-management as ideal types and understand that the continuum between
these two ends is occupied by organizational structures that borrow some characteris-
tics from both ends of the spectrum (see especially Lee and Edmondson 2017 Figure 1).
Martela Journal of Organization Design (2019) 8:23 Page 17 of 23

It has been argued that bureaucracy is “a matter of degree, rather than kind”(Hall
1963, p. 37) and behind the “static ideal type”there is “a living, changing, and diverse
set of practices”(Alvesson and Thompson 2006, p. 501). These ideal types illuminate
what kind of organizational configurations are possible and what solutions to the key
problems of organizing tend to co-occur, thus illuminating and helping to explain how
various real-life organizations function. In trying to explain such organizational outliers
as Morning Star, Buurtzorg, or Valve, the explanatory power of the bureaucratic and
adhocratic ideal types seems not to capture the structural arrangements and dynamics
of these organizations. The SMO as an ideal type thus helps us to understand this rela-
tively novel form of organizational design.
Self-organizing as an ideal type of organizational structure seems to have many at-
tractive qualities. It promises to offer employees more individual-level autonomy and
organizational-level decision-making power. Thus, at best such organizations could an-
swer calls for increased employee emancipation and empowerment. Organizations
practicing self-management have often experimented with a plethora of interesting new
practices ranging from transparent salaries to bottom-down coordination of operations,
making them interesting contexts for case studies. If we would have rigorous evidence
of their success (or failure) in increasing employee engagement, emancipation or prod-
uctivity, this could pave the way for a more wide-scale adaptation of such practices.
At the same time, there is a clear need for research looking critically at self-
management and whether it really delivers on what it promises. Popular press tends to
write uncritical success stories where ‘the revolution’brings nothing but good for both
the employees and the bottom line. The reality is, naturally, more complicated (Parker
and Parker 2017). Self-management can also bring forth confusion, stress, and chaos.
When Zappos adapted full-scale holacracy in 2015, their annual turnover rate increased
from roughly 20% to roughly 30% (Lam 2016). Accordingly, at the organizational level,
we need research on the real consequences and potential downsides of self-
management. What are the pitfalls and challenges that need to be taken into account
in striving towards more self-management? Which groups benefit from it, which do
not? For example, having explicit performance criteria can help various minorities ad-
vance in organizations. Given the various well-known social and cognitive biases of us
humans, the more informal structures in self-managing organizations can in the worst
case lead to social interaction and recruitment to projects being based on homophily
(Felin 2015), at the expense of various minorities. The problem is made more acute be-
cause the informal power structure might make it harder to detect such a bias. While
in a traditional bureaucracy it is relatively easy to count the gender ratio in top or mid-
dle management, the distribution of informal power within SMOs might resemble a
high-school popularity contest (see Warr 2013), where being similar to the CEO or top
management in terms of gender, age, or hobbies might yield significant power without
it being easy to pinpoint. Thus, to avoid such biases that can easily emerge without
anyone consciously driving them, SMOs probably need to have explicit policies in place
that counter such problems. Furthermore, there is no reason why “being overruled (or
even ostracized) by a community of peers”would be less bad for the individual than
“being overruled by a boss”(Puranam and Håkonsson 2015, p. 4), and thus the commu-
nity norms and ways of handling disputes can significantly affect the psychological ex-
perience of being part of an SMO.
Martela Journal of Organization Design (2019) 8:23 Page 18 of 23

By offering a supposed antidote to hierarchical structures, self-management can also
offer important insights into research on hierarchies and their functions in organiza-
tions. It has been argued that the pervasiveness of hierarchies is due to the fact that in
most contexts they are adaptive and can facilitate organizational success (Magee and
Galinsky 2008; Halevy et al. 2011). On a psychological level, hierarchies structure work
making it more predictable and thus often give employees a sense of order and security
(Pfeffer 2013). Self-management can enrich our understanding of hierarchies and per-
haps demonstrate the contexts and requirements where such traditional hierarchies are
not needed. At the same time, self-managing organizations might give rise to informal
hierarchies leading one to ask whether “hierarchy”is a monolithic concept or whether
there can be several different types of hierarchies (cf. Hardy and Leiba-O’Sullivan
1998), some more adaptive than others. For example, dominance and prestige have
been shown to be two distinct paths to having influence and power over others (Ander-
son and Kilduff 2009; Cheng et al. 2013). While formal hierarchies emphasize domin-
ance, SMOs that lack such hierarchies might still have significant differences in regard
to how much prestige-based power various individuals have within the organization.
Some people, due to their knowledge and expertise might exercise disproportional
power within the organization without necessarily having any formal credentials.
Whether or not such prestige-based power distances are good or bad for organizations
is an important question, and SMOs would provide a particularly apt context to study
it.
SMOs would also provide an interesting context to study strategy. Research on strat-
egy has sometimes been skeptical about the success of bottom-up strategy formulation
due to, for example, in-group biases within organizational subunits (Reitzig and Soren-
son 2013). Can such bottom-up strategy formulation lead to bold enough decisions, like
entering a new market? Or does it rather lead to organizational inertia? There is a risk
that people within the organization are too invested in their own sub-task to be able to
see when that sub-task as a whole has become a burden for the organization and its
goals. It is hard to imagine strategic decisions like abolishing or outsourcing a company
division being made by people within that division. It is thus perhaps no surprise that
in their examination of four SMOs, Lee and Edmondson (2017) note that firm strategy
is a dimension that tends not to be decentralized even within these organizations. Thus
it could be the case that strategy is one of the dimensions that resist decentralization
even within SMOs. Yet, it would be interesting to study strategy formulation in SMOs
to see whether the process within them is more inclusive and whether one could iden-
tify examples of SMOs in which even the strategy formulation has been to a significant
degree decentralized.
At the individual level, it has been argued that self-management requires ‘more’from
the employees as compared with traditional organizations in the sense that in a bureau-
cracy many functions such as scheduling, goal-setting, and prioritizing are often taken
care of by the supervisor (Martela and Kostamo 2017). Here we were able to identify
several reasons why more is required from the employees. Being responsible to identify
what tasks need to be done to advance organizational goals (task division) means that
employees must exhibit more proactivity and personal initiative than in traditional or-
ganizations where it is enough to execute tasks allocated to oneself. Given the lack of
monitoring (rewarding desired behavior), employees need to find motivation from
Martela Journal of Organization Design (2019) 8:23 Page 19 of 23

within, leading to increased expectations of intrinsic motivation and engagement. As
supervisors are not there to resolve conflicts (eliminating freeriding), employees must
have considerable skills in conflict resolution and confronting colleagues not fulfilling
their tasks. To make bigger decisions in a way that benefits the organization as a whole
(providing direction), they need to be able to understand the organizational wholeness,
leading to requirements to be able to, e.g., understand the financial realities of the
company.
Accordingly, self-managing organizations offer an especially illuminating context to
study concepts such as self-leadership (Neck and Houghton 2006; Plowman et al.
2007), proactivity (Grant and Ashford 2008), self-determination, and intrinsic motiv-
ation at work (Deci et al. 2017; Martela and Kostamo 2017). Furthermore, the lack of
formal supervisors makes this an interesting context to study leadership, especially vari-
ous forms of distributed leadership such as shared leadership (Carson et al. 2007), dis-
persed “power-with”leadership (Salovaara and Bathurst 2018), relational leadership
(Uhl-Bien 2006), and leadership of emergence (Lichtenstein and Plowman 2009). These
leadership theories argue that leadership does not have to be concentrated in formal
leadership positions, and given the lack of such formal positions, SMOs could offer im-
portant insight about which persons, routines, structures, or systems take care of the
functions and tasks that have traditionally been taken care of by supervisors.
It needs to be acknowledged that this article has focused on organizational structures,
which are rather general design features of how an organization has structured their
authority relations, their decision-making powers, information-distributing channels,
and similar. Organizations also have other important dimensions such as organizational
routines (e.g., Feldman and Pentland 2003) or organizational culture (e.g., Schein 2010)
that have not been covered by this review. An important task for future research would
be to examine what implications for organizational culture SMOs have. For example,
what cultural aspects are necessary for SMOs to function or are some other aspects of
culture incompatible with the structures that SMOs rely on? Similarly, what kind of
routines are beneficial or detrimental in the case of SMOs? Examining these questions
will help us to get an even broader picture of the nature of SMOs.
SMOs also offer an interesting case for Adler’s(2001) three organizational logics:
markets, hierarchies, and communities. While in markets, affiliations are handled
through market contracts and in hierarchies through employment contracts, in com-
munities the membership is looser and more trust-based (Kolbjørnsrud 2018). In that
sense, SMOs would be categorized as hierarchies as the employees are affiliated to the
organization through contracts. Furthermore, in SMOs the roles, responsibilities, and
expectations are typically explicit and involve (often peer-based) mechanisms of enfor-
cing them, rather than tacit and based on gifts, favors, and fluctuating levels of involve-
ment as in communities (Adler et al. 2008). At the same time, communities are
characterized as places where “actors self-organize guided by shared values, rules, and
protocols”(Kolbjørnsrud 2018, p. 3), a characterization that seems to fit well with
SMOs. Thus SMOs seem to represent a way of introducing aspects of community logic
within employment-based organizations while retaining certain aspects of the ‘hierarch-
ical’logic, although implemented in less-hierarchical ways. Thus, when observed from
the point of view of these three logics, SMOs could be characterized as a hierarchy-
community hybrid (see Kolbjørnsrud (2018) who sees holacracies as such hybrids).
Martela Journal of Organization Design (2019) 8:23 Page 20 of 23

Conclusion
”Management is the least efficient activity in your organization. …A hierarchy of man-
agers exacts a hefty tax on any organization.”Thus proclaims Gary Hamel (2011,p.
50), predicting that due to the flexibility, innovativeness, and cost benefits associated
with self-organizing, we are going to see much more such organizations in the future.
Is this the future we are heading for?
It is too early to say, despite the encouraging examples of some pioneering compan-
ies. The “end of bureaucracy”has been proclaimed many times ever since Weber, but it
still arguably is the dominant form of organizing corporations and public institutions
(Alvesson and Thompson 2006). However, given that more volatile business environ-
ments require an ability to adapt fast to changing environments, and a more profes-
sionalized workforce requires less managerial oversight, time could be right to move
towards more self-management in many sectors. Furthermore, sophisticated IT systems
have made it possible to coordinate the actions of thousands of people in real-time
making self-management more possible than in previous eras. Accordingly, the prime
examples of self-organizing such as Zappos, Buurtzorg, Morning Star and others (see
Laloux 2014) might not be only isolated examples but represent the dawn of a new era
of organizational structure. While certainly not suitable for all industries and all con-
texts, there arguably are many organizations today that would benefit from taking one
or more steps towards self-organizing. However, if we want to see such a shift towards
self-managing organizations, this only becomes possible once we’ve identified what are
the key enabling structures that need to be in place for such organizations to work.
Building this understanding is what the present work has attempted to contribute to.
Abbreviations
SMO: Self-managing organization
Author’s contribution
The author wrote the manuscript, read and approved the final version.
Funding
Funding for this project was provided by a grant from Business Finland (MODe –Minimalist Organizational Design).
Availability of data and materials
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.
Competing interests
The authors declare that they have no competing interests.
Received: 23 November 2018 Accepted: 12 November 2019
References
Adler PS (2001) Market, hierarchy, and trust: the knowledge economy and the future of capitalism. Organ Sci 12:215–234
Adler PS (2012) The sociological ambivalence of bureaucracy: from Weber via Gouldner to Marx. Organ Sci 23:244–266
Adler PS, Kwon S-W, Heckscher C (2008) Perspective—professional work: the emergence of collaborative community. Organ
Sci 19:359–376
Alvesson M, Thompson P (2006) Post-bureaucracy? In: Ackroyd S, Batt R, Thompson P, Tolbert PS (eds) The Oxford Handbook
of Work and Organization. Oxford University Press, Oxford, pp 485–507
Anderson C, Kilduff GJ (2009) The pursuit of status in social groups. Curr Dir Psychol Sci 18:295–298
Astley WG, Sachdeva PS (1984) Structural sources of intraorganizational: power: a theoretical synthesis. Acad Manage Rev 9:
104–113. https://doi.org/10.5465/amr.1984.4278071
Barker JR (1993) Tightening the iron cage: Concertive control in self-managing teams. Adm Sci Q 38:408–437
Bernstein E, Bunch J, Canner N, Lee M (2016) Beyond the holacracy hype. Harv Bus Rev:38–49
Bernstein ES (2012) The transparency paradox: a role for privacy in organizational learning and operational control. Adm Sci
Q 57:181–216. https://doi.org/10.1177/0001839212453028
Bernstein ES (2017) Making transparency transparent: the evolution of observation in management theory. Acad Manag Ann
11:217–266
Beste Werkegevers (2017) Eerdere winnaars
Martela Journal of Organization Design (2019) 8:23 Page 21 of 23

Burton RM, Håkonsson DD, Nickerson J et al (2017) GitHub: exploring the space between boss-less and hierarchical forms of
organizing. J Organ Des 6:1–19. https://doi.org/10.1186/s41469-017-0020-3
Carson JB, Tesluk PE, Marrone JA (2007) Shared leadership in teams: an investigation of antecedent conditions and
performance. Acad Manage J 50:1217–1234. https://doi.org/10.2307/20159921
Cheng JT, Tracy JL, Foulsham T et al (2013) Two ways to the top: evidence that dominance and prestige are distinct yet
viable avenues to social rank and influence. J Pers Soc Psychol 104:103–125
Cohen SG, Ledford GE (1994) The effectiveness of self-managing teams: a quasi-experiment. Hum Relat 47:13–43
Deci EL, Olafsen AH, Ryan RM (2017) Self-determination theory in work organizations: the state of a science. Annu Rev Organ
Psychol Organ Behav 4:19–43. https://doi.org/10.1146/annurev-orgpsych-032516-113108
Deci EL, Ryan RM (2000) The“what”and“why”of goal pursuits: human needs and the self-determination of behavior.
Psychol Inq 11:227–268
Ezzamel M, Willmott H (1998) Accounting for teamwork: a critical study of group-based systems of organizational control.
Adm Sci Q 358–396
Fehr E, Fischbacher U, Gächter S (2002) Strong reciprocity, human cooperation, and the enforcement of social norms. Hum
Nat 13:1–25
Fehr E, Gächter S (2002) Altruistic punishment in humans. Nature 415:137–140
Feldman MS, Pentland BT (2003) Reconceptualizing organizational routines as a source of flexibility and change. Adm Sci Q
48:94–118. https://doi.org/10.2307/3556620
Felin T (2015) Valve corporation: strategy tipping points and thresholds. J Organ Des 3:10–11
Fiss PC (2007) A set-theoretic approach to organizational configurations. Acad Manage Rev 32:1180–1198
Foss NJ (2003) Selective intervention and internal hybrids: interpreting and learning from the rise and decline of the Oticon
spaghetti organization. Organ Sci 14:331–349
Galbraith JR (1973) Designing complex organizations. Addison-Wesley, Reading, Ma
Gelles D (2015) At Zappos, pushing shoes and a vision. N Y Times 136
Grant AM, Ashford SJ (2008) The dynamics of proactivity at work. Res Organ Behav 28:3–34
Grant AM, Gino F, Hofmann DA (2011) Reversing the extraverted leadership advantage: the role of employee proactivity.
Acad Manage J 54:528–550
Gray BH, Sarnak DO, Burgers JS (2015) Home care by self-governing nursing teams: the Netherlands’Buurtzorg model.
Commonw. Fund Pub 1818 14:1–10
Grote G, Guest D (2017) The case for reinvigorating quality of working life research. Hum Relat 70:149–167. https://doi.org/10.
1177/0018726716654746
Hackman JR, Oldham GR (1980) Work redesign. Addison Wesley, Reading, Ma
Hakanen JJ, Perhoniemi R, Toppinen-Tanner S (2008) Positive gain spirals at work: from job resources to work engagement,
personal initiative and work-unit innovativeness. J Vocat Behav 73:78–91. https://doi.org/10.1016/j.jvb.2008.01.003
Halevy N, Y. Chou E, D. Galinsky A (2011) A functional model of hierarchy: why, how, and when vertical differentiation
enhances group performance. Organ Psychol Rev 1:32–52
Hall RH (1963) The concept of bureaucracy: an empirical assessment. Am J Sociol 69:32–40
Hamel G (2011) First, let’s fire all the managers. Harv Bus Rev:48–60
Hamel G (2014) Bureaucracy must die. Harv Bus Rev Retrieved from https://hbr.org/2014/11/bureaucracy-must-die
Hamel G, Zanini M (2016) Excess management is costing the U.S. $3 trillion per year. Harv Bus Rev
Hardy C, Leiba-O’Sullivan S (1998) The power behind empowerment: implications for research and practice. Hum Relat 51:
451–483
Kolbjørnsrud V (2018) Collaborative organizational forms: on communities, crowds, and new hybrids. J Organ Des 7:1–21.
https://doi.org/10.1186/s41469-018-0036-3
Laloux F (2014) Reinventing organizations: a guide to creating organizations inspired by the next stage of human
consciousness. Nelson Parker, Brussels
Lam B (2016) Why are so many Zappos employees leaving? The Atlantic
Lawrence PR, Lorsch JW (1967) Organization and environment: managing differentiation and integration. Harvard University,
Boston, Ma
Lee MY, Edmondson AC (2017) Self-managing organizations: exploring the limits of less-hierarchical organizing. Res Organ
Behav 37:35–58. https://doi.org/10.1016/j.riob.2017.10.002
Lichtenstein BMB, Plowman DA (2009) The leadership of emergence: a complex systems leadership theory of emergence at
successive organizational levels. Leadersh Q 20:617–630
Magee JC, Galinsky AD (2008) Social Hierarchy: the self-reinforcing nature of power and status. Acad Manag Ann 2:
351–398
March JG, Simon HA (1993) Introduction to the second edition. In: March JG, Simon HA (eds) Organizations (2nd ed.). Wiley,
New York, pp 1–19
Martela F, Kostamo T (2017) Adaptive self-organizing: The necessity of intrinsic motivation and self-determination. In: Eskola A
(ed) Navigating through changing times - knowledge work in complex environments. Routledge, New York
McGregor D (1960) The human side of the enterprise. McGraw-Hill, New York
Meyer AD, Tsui AS, Hinings CR (1993) Configurational approaches to organizational analysis. Acad Manage J 36:1175–
1195
Mintzberg H (1980) Structure in 5’s: a synthesis of the research on organization design. Manag Sci 26:322–341
Mintzberg H, McHugh A (1985) Strategy formation in an adhocracy. Adm Sci Q 30:160–197
Neck CP, Houghton JD (2006) Two decades of self-leadership theory and research: past developments, present trends, and
future possibilities. J Manag Psychol 21:270–295. https://doi.org/10.1108/02683940610663097
Parker S, Parker M (2017) Antagonism, accommodation and agonism in critical management studies: alternative organizations
as allies. Hum Relat 70:1366–1387. https://doi.org/10.1177/0018726717696135
Pfeffer J (2013) You’re still the same: why theories of power hold over time and across contexts. Acad Manag Perspect 27:
269–280
Plowman DA, Solansky S, Beck TE et al (2007) The role of leadership in emergent, self-organization. Leadersh Q 18:341–356
Martela Journal of Organization Design (2019) 8:23 Page 22 of 23

Puranam P, Alexy O, Reitzig M (2014) What’s“new”about new forms of organizing? Acad Manage Rev 39:162–180
Puranam P, Håkonsson DD (2015) Valve’s way. J Organ Des 4:2–4
Reitzig M, Sorenson O (2013) Biases in the selection stage of bottom-up strategy formulation. Strateg Manag J 34:
782–799
Robertson BJ (2016) Holacracy: The revolutionary management system that abolishes hierarchy. Penguin
Salanova M, Schaufeli WB (2008) A cross-national study of work engagement as a mediator between job resources and
proactive behaviour. Int J Hum Resour Manag 19:116–131
Salovaara P, Bathurst R (2018) Power-with leadership practices: an unfinished business. Leadership 14:179–202
Schein EH (2010) Organizational culture and leadership. John Wiley & Sons, San Francisco, CA
Simon HA (1991) Organizations and markets. J Econ Perspect 5:25–44
Trist EL, Bamforth KW (1951) Some social and psychological consequences of the Longwall Method of coal-getting. Hum
Relat 4:3–38
Uhl-Bien M (2006) Relational leadership theory: exploring the social processes of leadership and organizing. Leadersh Q 17:
654–676
Warr P (2013) Valve’s flat management structure “like high school.”Wired UK
Weber M (1946) Bureaucracy. In: Gerth HH, Mills CW (eds) Essays in sociology. Oxford University Press, New York, pp 196–244
Weber M (1968) Economy and society: an outline of interpretive sociology. University of California Press, Berkeley
Publisher’sNote
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Martela Journal of Organization Design (2019) 8:23 Page 23 of 23