2.1.1.1. Contingent view on coordination

There are many authors who reflect on “how” activities and people are coordinated within teams. They have developed a set of typologies that identify the most appropriate coordination mechanisms with regard to action contexts. For the proponents of this (still) dominant contingency view (for example, [MAR 58, LAW 67, THO 67, VAN 76] and [MIN 78]), the choice of mechanisms is a matter of “adjustment – or furthermore of strategic alignment” [PIC 02] in relation with the contextual constraints and opportunities, external and/or internal.

This perspective synthetically identifies two main contingency factors that affect coordination: the level of interdependency between parties [THO 67, CHE 83, MAL 94, CRO 97, GIT 02] on one hand, and the degree of environmental uncertainty [MAR 58, GAL 73, VAN 76, GUP 94], on the other hand. Teams and organizations are called upon to develop their coordination mechanisms in relation to these contingency elements. According to Okhuysen and Bechky [OKH 09], certain contributions are integrated into contingency effects by acting at the level of operating work stations and modes (for example, [TAY 16] and [CHA 62]), while others focus more on the conception of management systems dedicated to structures, roles and rules (for example, [FAY 49, THO 67] and [MIN 78]). Despite these divergences, all the authors agree on the following principle: whether operational or structural, the arrangements are conceived in order to process a volume of information corresponding to the number and complexity of tasks to be accomplished.

This is how March and Simon [MAR 58] have come to consider the plan an effective formal coordination mechanism when the organization operates in a routine context, where information is redundant and reliable, while retroaction would prove more adapted to unexpected and sudden situations. Thompson [THO 67] considers that, faced with high uncertainty, teams are coordinated effectively when they use flexible mechanisms, such as mutual adjustment and informal interactions. Along the same lines, Mintzberg [MIN 78] defends the idea of a continuum between five mechanisms of coordination, where increasing complexity of the situations induces “relaxation” of coordination mechanisms (from formal mechanisms, such as direct supervision, to more informal mechanisms, such as mutual adjustment).

The contingent approach thus considers that formal and informal coordination mechanisms coexist in organizations. The teams either substitute one for the other, or they superimpose them, depending on their needs. Substitution refers to the use of one mechanism instead of another, while superimposition refers to a form of mechanism accumulation, which the teams will use as a source of appropriate elements, according to the situation [GOD 14]. The major contribution of the contingent approach lies, therefore, in the differentiation of coordination mechanisms and the analysis of how they are used by the actors.

2.1.1.2. Mechanisms, means and tools of coordination

The question of the nature and aims of coordination mechanisms remains open in the contingent literature. As noted by Alsène and Pichault [ALS 07], there is a tendency to qualify the set of coordination elements mobilized to coordinate tasks and people within organization and teams as “mechanisms”.

Even so, there are significant differences between the elements of coordination. A thorough review of the literature makes a distinction: the actors use at the same time mechanisms, means and tools of coordination in their daily activities (Table 2.1).

Coordination mechanisms are those typically referred to by contingency theories, namely as the “most basic elements […] that hold organizations together” [MIN 78]. Mutual adjustment, considered a process of informal communication between actors [MIN 78], differs from relational coordination [GIT 02] to the extent that the latter focuses more on the role played by social networks in support of coordination. In this context, Gittell [GIT 02] takes into consideration both the informal dimension of interactions and the power of social relations and networks [GRA 85].

The means of coordination refer to devices that individuals use to reach their objectives. They diverge from the tools, which are instruments of coordination, namely the support available for implementing and facilitating coordination. In the contingency view, the tools refer to formal and informal artifacts that support coordination. In this context, information and communication technology (ICT) plays a central role.

The contributions rooted in the classical perspective approach technologies as technical and computer tools in the service of mechanisms and means of coordination already existing in the organization [CAB 99]. The contingency factors most frequently stressed are the degree of task interdependence, the size of teams and the volatility of the action context. These factors would permit management authorities to identify the type of technologies to implement [KIM 88, KEL 94], depending on their functionalities and expected effects.

In this context, the concept of fit – alignment or adequacy – between organizational context and technological artifact is frequently mentioned in literature [KIM 88, KEL 94, SHE 04, KOT 08]. For the most part, the theories of fit [GOO 95a, GOO 95b, JUN 08, GEB 09] explore the adequacy of technology, the task completed by the user and the context in which the task is executed in order to improve individual and collective work performances [GOD 13]. The answers to the question of “how” technologies contribute to proper team coordination propose measurements of how activity is affected by volume, quality and modes of information transmission. The contingency factors retained are task predictability, analyzability and coupling [KIM 88, KEL 94], group size [KIM 88, KIM 90], frequency of internal and external changes as well as the more or less routine nature of the task to be executed [KEL 94, SHE 04, CHE 08].

Rich as it may be, we have to admit that contingency literature offers few ideas for understanding how teams are concretely coordinated in a given situation. The works describe a range of coordination elements that need to be identified beforehand and implemented in order to adapt the work modes and/or the structures to contingency factors. These contingency factors determine which elements of coordination are most appropriate. In this context, actors intervene in the downstream coordination: they are approached as simple “users” of mechanisms, means and tools provided to them by the organization.

2.1.2.1. The notion of coordination solution

Relying on this knowledge base, Alsène and Pichault [ALS 07] propose the notion of “coordination solutions” in an attempt to account for how individuals work together (implement coordination mechanisms, means and tools). The authors perceive the coordination solution in terms of a “prescription proposed to (or imposed on) a group of employees, which tends to create coherence of their efforts”. Their nature depends on the work situations the teams are confronted with.

Alsène and Pichault [ALS 07] identify 11 work situations. For example, shift work describes a case of one or several employees who take over from one or several other employees; concurrent work corresponds to a situation where several employees participate simultaneously, but each on his or her own, to produce a collective output; or team work refers to the case of several employees working simultaneously, in a collective manner, to produce a collective output.

Each of these work situations has a typical corresponding coordination solution. As a result of research conducted in four large industrial companies, the authors have listed 15 solutions. For example, “operating procedures to follow” refers to a solution where actors have to use certain procedures known to all in order to complete their tasks; “assignments to take on” refers to cases where employees are asked to change their area of responsibility and/or to complete tasks pertaining to another job; “mandates to accomplish” denotes a coordination solution where individuals carry out an activity that does not officially belong among their tasks, but is part of their area of competences.

Using the notion of coordination solution, Alsène and Pichault [ALS 07] question some of the foundations of contingency theories, notably the idea developed by Mintzberg [MIN 79] according to which every organizational configuration corresponds to a coordination mechanism that dominates the mobilized continuum. The authors then demonstrate that approaching coordination at the level of work situation instead of that of structure broadens the analysis perspectives. In this case, the various work situations coexisting in the organization refer to 15 coordination solutions implemented by the actors, a fact that invalidates the principle of dominant mechanism.

However, the discussion around coordination “solutions” cannot solve the problem of limitations of contingency theories when the subject being studied is coordination in the extreme environment. In effect, in such a context teams navigate continuously between routine situations, marked by standardization and formalism, and unexpected situations that require flexibility and responsiveness. The pace of changes is high and random. In these conditions, to predetermine and impose coordination mechanisms or solutions becomes a complex, even counterproductive issue. As we will see further on, teams do not follow a logic of substitution or superimposition of coordination mechanisms or solutions designed upstream by the management [GOD 14]. They “produce” coordination on a daily basis.

2.1.2.2. Limitations of the classical perspective of coordination

There are four main limitations of the classical perspective on coordination [BOU 11]:

• – First, contingency theories are part of a post-Coasian representation, wherein the company is viewed as an information processor [COH 99]. This view is based on the idea of an environment that, even if marked by complexity, remains predictable enough to make possible the upstream designation and implementation of coordination mechanisms or solutions best adapted to various contingency factors. Alsène and Pichault [ALS 07] do not escape this first critique. As they qualify the coordination solutions as “prescriptions”, they implicitly relate them to a top-down view of coordination, where the manager and/or the organization have the capacity and means to identify upstream problems and advance to (or impose on) employees the effective solutions.

But, the passage from the industrial paradigm to the knowledge-based economy has radically changed work situations and organizational processes [PES 02]. Due to the complexity of action contexts, the upstream process whereby effective mechanisms or solutions are identified may become very expensive, if not impossible. Moreover, these mechanisms or solutions may emerge simultaneously with the development of the problem to be solved. The process through which coordination solutions are constructed may, therefore, prove to be bottom-up rather than top-down. In this context, it is very difficult to achieve preidentification of interdependences between work situations and coordination solutions, mechanisms, means and tools;

• – Second, the contingent perspective relies essentially on a “design” view of coordination [OKH 09], the envisaged coordination mechanisms referring to various arrangements: either at the level of structural devices to be implemented, such as activity planning (mechanistic system) in a slow evolving context; or at the level of forming transversal groups (organic structure) that aim to facilitate lateral exchanges in a more unstable context. Under these circumstances, coordination is approached at too aggregated a level to allow for comprehension of the phenomenon in all its complexity, and the hidden part of the iceberg remains unknown;
• – Third, the contingent models do not take into account the time dimension of coordination. It is not approached in terms of process, but only in terms of content (mechanisms, means and tools). The diachronic aspect of coordination is essential for understanding changes that organizations go through and the actors’ capacity to adapt [BOU 10, BOU 12, NIZ 12, GOD 14]. Team coordination is a continuous and situated process: continuous because the search for coordination takes place in real and historic time, stressing the irreversible and uncertain dimension of time (in [BER 46] Bergson speaks of “continuous creation of unforeseeable novelty”); situated because it evokes a social construct which emanates from interactions and interdependences between competences, expertise and specific knowledge [BEC 06, XIA 07, RIC 08] within a given context [FAR 06];
• – Finally, the level of granularity retained by the contingency theories is adequate at the scale of the organization or work unit. For example, they refer to large organization samples. This is the case for the contribution of Burns and Stalker [BUR 61] which studies around 20 cases of British companies in order to distinguish between mechanistic and organic structures. However, such analyses do not allow for an examination of how coordination is produced at the level of actors’ play and interactions, or for the opening of “this black box that contingency theorists are rather inclined to keep closed” [NIZ 12]. As these two authors note, the examination of individual and collective actions of individuals involved in coordination is possible only if we concentrate on microsociological situations. This research path looks then particularly attractive.

2.2.1.1. Drawing inspiration from contemporary sociology

Though recent, the practice turn in management science enjoys growing success. A good illustration is the strategy as a practice trend, which has become well known over the last 15 years (see, for example, [WHI 06] and [JAR 07]). But as Whittington [WHI 06] stresses, strategy is not the only specialist area influenced by the practice turn. For example, contributions of Orlikowski and Yates [ORL 94] and Orlikowski [ORL 00] to the management of information systems, those of Brown and Duguid [BRO 91] or Wenger [WEN 00] to the understanding of organizational learning dynamics or the works of Brownlie and Heweren [BRO 11] in marketing show that the practice-based approach is a source of inspiration for many specialized areas of management.

This practice turn is naturally rooted in contemporary sociological theories [GHE 06]. In particular, the close intertwining of field, habitus and practice concepts is in Bourdieu’s view one of the main sociological bases of this approach. Starting from a critique of intellectualist models, Bourdieu develops a sociology of action leading to the approach of practice as “the dialectic of the opus operatum and the modus operandi, of the objectified products and the incorporated products of historical practices; of structures and habitus” [BOU 80]. Practices are thus dispositions to act that one is not conscious of. They are the product of habitus, in the sense of a directory of “principles of generation” [BOU 80], of “durable, transposable dispositions” [BOU 80] acquired through early and adult life experiences. Habitus in its turn is part of one or several fields, which are spheres of social life organized over time. Practices develop in and through action, under the influence of habitus. This consideration leads Bourdieu to the introduction of the social agent. The social agent’s actions “are both inwardly and outwardly directed” [BON 97]. According to Bourdieu, social action reflects social structures; structures have a primacy over action.

The contributions of Giddens [GID 84], and his concepts of action and structure, are also frequently quoted by authors who share the practice-based approach, particularly in the field of management of information systems. In parallel with the works of Bourdieu, Giddens develops his theory of structuration insisting more on the dialectical dynamics of social structures and action. Here, Giddens prefers the term “actor” instead of that of social agent used by Bourdieu. Giddens’s actor is competent, capable of controlling and directing his or her actions. Controlled actions are for him or her a source of new knowledge that he or she takes into account in order to act. The action–actor link is considered recursive, namely based on a circular dynamics that involves repetitive process.

The second key concept of Giddens’s thought is that of structure. Structure evokes a virtual order consisting of a set of rules and resources that actors use in their actions. Thus, they participate in a recursive manner in the development of a social system (for example, a society or an organization). Here, recursivity characterizes retroactions between structure (consisting of rules and resources) and actions: structure directs (controls) action just as much as action acts on structure. Structure represents the means and result of the action it recursively organizes at the same time [ROJ 00]. Giddens uses the term “duality of structure” [GID 84] to describe this phenomenon of mutual transformation of structure and action. It produces and reproduces the structural properties of the social system: certain structural properties rely on rules that can contribute to sense-making. When these rules constitute shared interpretative diagrams that permit actors to communicate, Giddens speaks of structures of signification. Other structural properties govern the power relations between actors. The author qualifies them as structures of domination. Finally, certain structural properties consist of norms, moral codes, conventions that embody the established order at a given moment in the evolution of the social system. Giddens then speaks of structures of legitimation.

Through actions and interactions, the structures of signification, domination and legitimation become institutionalized to the extent that they enter a process of structuration and become durable. For this reason, they affect at their turn the actors’ actions and interactions, either as constraints or as “empowering” factors.

2.2.1.2. From practice to practices

The practice-based approach refers to the analysis of microactions through which actors organize activity [SCH 01, ORL 02, GHE 06]. It therefore offers the opportunity to grasp the continuous and situated character of coordination. Such a perspective leads the researcher to explore “the internal life” [GOL 06] of management situations, to understand how coordination is concretely produced when actors work in a routine environment and are suddenly confronted with unexpected situations. Thus, the level of analysis proposed by the practical perspective is microscopic. This leads in particular to an examination of individual and collective modes of action, interdependences and technological uses. In this last case, we will see hereafter that uses are practices. Uses translate the paths of users’ appropriation of technologies; they represent the way in which users implement them (“enact” them).

The authors who adopt the practice-based approach insist that microactions cannot be approached independently of the social foundations they are shaped by. Thus, they distinguish practice – or praxis – from practices (for example, [REC 02, WHI 06] and [JAR 07]). “The practice” describes the interdependences between individual actions, the flow of activities in situation [JAR 07], as well as the institutional environment in which they are expressed. In that respect, practice is “a mode, relatively stable in time and socially recognized, of ordering heterogeneous items into a coherent set” [GHE 06].

As for “the practices”, they refer to a “routinized type of behavior which consists of several elements, interconnected to each other: forms of bodily activities, forms of mental activities, ‘things’ and their use, know how, states of emotion and motivational knowledge” [REC 02]. Practices thus refer to the actors’ way of doing things in the situation. They represent organized (or more precisely “organizing”) actions that are part of a context of specific social relations, rules of behavior and ways of doing things. In this respect, practices partake of the achievement of objectives that give them meaning. They are “intrinsically connected to ‘doing’ because they provide the behavioral, cognitive, procedural, discursive and physical resources through which multiple actors are able to interact in order to socially accomplish collective activity” [JAR 07].