Abstract
Coordination among the teams of a multiteam system is necessary in order to initiate and maintain inter-team interdependence. In turn, coordinated interdependence is required if the teams in a multiteam system are to work together toward a common outcome and succeed as an organized entity. A literature review indicates that multiteam research has indicated that three basic coordination mechanisms—mutual adjustment, direct supervision, and standardization—are used to coordinate interdependence among teams. The review also reveals that multiteam systems research has seldom examined transitions among inter-team coordination mechanisms and has rarely investigated precipitants that trigger mechanism transitions. In light of this finding, this article describes theorized transitions and identifies precipitant factors likely to stimulate these transitions. It concludes that transitions and precipitants merit significant attention in future multiteam systems research in order to render a more complete understanding of inter-team coordination.
As work has become more complex, dynamic, and demanding, work organizations have turned to multiteam systems—organizational units composed of multiple teams—to accomplish tasks that individuals or teams could not complete if working alone (Marks et al., 2005; Mathieu et al., 2017; Shuffler et al., 2015). Interdependence among the teams of a multiteam system (MTS) is fundamental to MTS success, enabling teams to share knowledge, effort, and resources in the pursuit of common objectives (Mathieu et al., 2001). Coordination, in turn, is necessary to create and sustain interdependence among MTS teams by supporting communication among teams and synchronizing inter-team efforts (e.g., Davison et al., 2012).
MTS research has noted the importance of inter-team coordination and a significant stream of research has examined different mechanisms used to coordinate work among MTS teams. As this stream has developed, researchers active within it have often used differing theoretical models and varying conceptual terminology while studying the same basic coordination processes. For example, what has been called leadership in some studies and examined using the theoretical lens of leader effectiveness has been referred to as hierarchy in others and theorized as a structural process (e.g., Davison & Hollenbeck, 2012; DeChurch, 2002). Consequently, research results have resisted cumulation, and conclusions bridging conceptual differences remain to be discovered.
The purpose of the present article is to provide the necessary conceptual bridge and aggregate research results in order to formulate summary conclusions and identify pathways for future research on MTS coordination. Using a model developed by Henry Mintzberg (1979, 1993) as a theoretical frame, a literature review will standardize the terminology used in MTS coordination research and verify that MTS studies have shown that inter-team coordination is accomplished using mutual adjustment, direct supervision, or standardization, the three types of basic coordination mechanism identified in Mintzberg's model. In addition to identifying different types of coordination mechanisms, Mintzberg's model also describes transitions among the types, suggesting precipitants that can lead one mechanism to lose efficacy and another to gain utility in its place. The literature review will indicate that MTS research has been less thorough in studying these transitions and precipitants. As a result, MTS research has not yet developed specific explanations that indicate how and why transitions occur. Contemporary understanding of inter-team coordination is incomplete, and actions founded on this understanding lack necessary empirical support.
Based on this finding, attention in the article will turn to an examination of transitions among inter-team coordination mechanisms and precipitants activating those transitions. Discussion will suggest that precipitants such as the number of teams requiring coordination, the types of interdependence shared among teams, and the degree of cognitive diversity within and among teams will render some coordination mechanisms ineffective and others efficacious. The article will conclude that research on transitions among basic coordination mechanisms and related precipitant factors is required if studies of inter-team coordination are to yield substantiated conceptual understanding and pragmatic managerial advice.
Why inter-team coordination is necessary and difficult
As defined by Mathieu et al. (2001), an MTS is a “team of teams” consisting of two or more component teams that join together to achieve complex or dynamic objectives. Teams of Red Cross volunteers from across the country that congregate to provide disaster relief are an MTS. So, too, are the academic departments of a business college that contribute to a shared curriculum. Each component team in an MTS is a distinguishable unit that may pursue its own distinctive goals and thus differs in basic characteristics from other teams in the MTS, but the teams in multiteam systems are also interdependent, sharing significant coupling in their pursuit of collective objectives and contributing productive resources toward the attainment of common or superordinate goals (Mathieu, 2012; Mathieu et al., 2001; Mathieu et al., 2018). Processes of inter-team coordination sustain this interdependence, merging individual and team actions and underlying resource support in ways such that teams work together to contribute to collective achievement (Marks et al., 2005; Mathieu et al., 2018; Scheerer et al., 2014). Inter-team coordination processes distribute knowledge of other teams’ actions and outcomes among interdependent teams, enabling them to organize activities, synchronize behaviors, and direct productive resources toward the achievement of shared outcomes (e.g., DeChurch & Marks, 2006; Marks et al., 2005). For example, police units from different local and federal agencies assigned the task of working together to provide metropolitan crowd control must communicate about how to position units, combine inter-unit efforts, and respond to changing conditions in order to produce an effective presence. Absent inter-team coordination, interdependence among teams weakens, organization diminishes, and the common direction required to achieve shared purpose deteriorates. Inter-team coordination is an essential process and necessary product.
Grouping members into teams within multiteam systems encourages intra-team coordination but discourages inter-team coordination (Lawrence & Lorsch, 1967; Melcher, 1976). Interpersonal propinquity within teams supports face-to-face communication and simplifies common supervision, both of which can be used to enact coordination among team members (Daft & Lengel, 1984). In addition, similarities that are the reason for grouping individuals together into teams, for example, similarities in what members do, what they produce, or where they produce it, also encourage coordination because they make it easier to understand other members’ actions and anticipate their work outcomes (Mintzberg, 1979, 1993). However, boundaries and separation between teams interrupt communication and impede collective supervision, and dissimilarities among teams in function, output, or location deter anticipation of outcomes and understanding of actions (March & Simon, 1958). Consequently, coordination among interdependent MTS teams is a greater challenge than coordinating within them. Inter-team coordination in multiteam systems is an important but especially difficult accomplishment.
A model of coordination mechanisms
Prior to the emergence of research on multiteam systems, coordination was identified in management and organization theory as both a practical challenge and a conceptual focus. In the resulting literatures, founding management principles recognized coordination as an essential management function (e.g., Fayol, 1930; Urwick, 1944), and benchmark organization theory identified coordination as a necessary underpinning of all forms of organization (e.g., Barnard, 1938; March & Simon, 1958). Theorists delineated a variety of mechanisms capable of enacting and sustaining coordination among interdependent units. For example, Thompson (1967) identified face-to-face communication, which he termed mutual adjustment, as an effective way to share coordinative information and synchronize individual behaviors and unit activities. March and Simon (1958) indicated that planning could be used to anticipate coordination needs and routinize coordinative behaviors. Van De Ven et al. (1976) identified impersonal, personal, and group coordination mechanisms and described differences in the degree and kind of influence associated with each type. Okhuysen and Bechky (2009) reviewed research on coordination in organizations and identified five mechanisms described in the literature: plans and rules, information objects and representations, roles, routines, and proximity.
From existing conceptualizations, Mintzberg (1979, 1993) identified three mechanisms that can enact coordination among interdependent parties, whether individuals or collectivities: mutual adjustment, direct supervision, and standardization. Mutual adjustment, as initially described by Thompson (1967), consists of face-to-face communication among interdependent parties (aircraft crewmembers in Thompson's research) in which information is exchanged informally in order to coordinate interdependent activities. Such information may include when, where, and how work will be handed off between interdependent individuals or groups. For example, the members of software development teams working on large multi-team projects coordinate among teams by talking with the members of other teams to reach agreement on issues such as anticipated production schedules or sequencing of ongoing work handoffs (Nyrud & Stray, 2017). Conversations occur as coordination problems emerge and are resolved through interpersonal communication.
Mutual adjustment requires that interdependent parties adapt to each other based on information shared in real-time. To satisfy this requirement, control of the work being performed must rest in the hands of the parties who are mutually adjusting. Mutual adjustment as envisioned by Thompson (1967) and Mintzberg (1979) also requires physical proximity. Consequently, geographic dispersal renders mutual adjustment ineffective. However, if the concept of mutual adjustment is broadened to include technologically enabled communication alternatives such as texting, phone messaging, or video conferencing, colocation is not required, and mutual adjustment can coordinate over distance. Technology effects such as this lie outside the boundaries of Mintzberg's model and are identified later in this article as a topic meriting additional theorizing and future MTS research.
Direct supervision, labeled coordination by line managers and unit supervisors by Van De Ven et al. (1976), is a second basic coordination mechanism and enacts coordination by having one party take responsibility for the work of others. In the case of inter-team coordination, this party may be an individual MTS manager or a leadership team composed of individuals from the various teams (and organizations, if relevant) being coordinated (e.g., Davison et al., 2012). With this responsibility comes the authority to decide which tasks will be performed, who will perform them, and how the outcomes of task performance will be integrated to produce coordinated performance. Direct supervision requires differentiation in the power granted to the supervisor versus the power allocated to subordinates, and thus creates and originates in hierarchy. Coordinative control in instances of direct supervision is transferred from the parties engaged in work to the supervisor who directs them (Gittell, 2001). A product development unit in design firm IDEO composed of several smaller creativity teams led by a central product manager exemplifies inter-team coordination by direct supervision (Brown, 2019).
Standardization, the third coordination mechanism identified by Mintzberg, achieves coordination by planning coordinative activities in advance and designing these activities into the work being done, before it is done (Hall et al., 1967; Pugh et al., 1963). In multiteam systems, standardization might include the set of basic procedures that rural firefighters plan in advance and use to establish coordination among units from different towns during joint responses to multiple alarm events. Questions of when, where, and how are answered beforehand and are encoded in standard operating procedures. Because coordination is incorporated into planned activities, reliance on real time communication can be reduced (March & Simon, 1958). In instances of coordination by standardization, the execution of standardized plans creates the coordination needed to sustain interdependence and achieve shared outcomes.
Mintzberg's model is a concise integration of prior theory and is regularly referenced in contemporary research (e.g., Hartnell et al., 2019; Joseph & Gaba, 2020; Raveendran, 2020). MTS researchers have identified the Mintzberg model as relevant to the study of inter-team coordination (e.g., Nyrud & Stray, 2017; Scheerer et al., 2014). It is particularly germane in the present instance due to its balanced treatment of mechanism types as well as transition processes and precipitant factors, therefore, its ability to organize the findings of existing MTS research and draw attention to areas of research that have yet to be investigated.
MTS research on inter-team coordination
Attention to the necessity of coordination among MTS teams is evident in the book chapter by Mathieu et al. (2001) that initiated current research on multiteam systems. Mathieu and colleagues indicated that interdependence—“mutual reliance, determination, influence, and shared vested interest in processes used to accomplish work activities” (2001, p. 7)—among component teams is an essential feature of multiteam systems. Building on this insight, MTS theory and analyses have investigated the use and effectiveness of different inter-team coordination mechanisms. Research has considered the options of lower-order mutual adjustment, examined as horizontal communication or informal information exchange among teams; direct supervision, studied as vertical communication, hierarchy, or leadership; standardization, assessed as standardization, planning, or the possession of shared interpretive perspectives; and higher-order mutual adjustment, captured in the coordinative function of liaisons, boundary spanners, and leadership meetings.
To integrate the findings of existing research on inter-team coordination and set the stage for a discussion of topics not yet examined, pertinent MTS research is reviewed in this section. To identify studies to include in the review, I conducted a search for multiteam systems theory and research on Google Scholar, ProQuest, and Web of Science between January 1981 (the year Mathieu and colleagues published the article that initiated MTS research) and July 2022. Search terms included “multiteam systems,” “multi-team systems,” multiteam, multi-team, and inter-team. I retained articles, book chapters, proceedings papers, and dissertations if their abstracts or introductions appeared to indicate that they addressed inter-team coordination methods or related multiteam processes. Next, I scanned the reference sections of all collected documents to identify additional documents not originally included, and I added newly identified documents to the review collection. From the resulting sample of 55 documents, I omitted 23 conceptual articles. I also removed six empirical papers that were found to be unrelated to the review's topic (e.g., Davison, 2012; Marks et al., 2005; Millikin et al., 2010). A final collection of 25 studies reported in 26 documents were included in the following review. Table 1 summarizes basic characteristics of the reviewed studies.
Study characteristics.
Lower-Order mutual adjustment
|Mintzberg (1979, 1993) indicated that a lower-order form of mutual adjustment, involving direct communication among interdependent parties, is the most basic type of coordination mechanism. Several MTS theorists have identified horizontal, face-to-face communication as an important means of achieving coordination among the teams in multiteam systems (e.g., Mathieu et al., 2001; Rico et al., 2018; Scheerer et al., 2014).
Empirical analyses of lower-order mutual adjustment include a study by Hoegl et al. (2004), who found that direct horizontal interaction between product development teams led to coordinated effort in multiteam projects and that this coordination enabled teams to complete their work on time. Mutual adjustment allowed the teams to synchronize technical details and connect work activities between teams. Nyrud and Stray (2017), in a second study, noted that large-scale software development procedures are especially challenging because they involve multiple teams and require extensive inter-team coordination. Findings of their study indicated that the most important and most frequently used inter-team coordination mechanism was informal mutual adjustment among project members from different teams. In a third study, De Vries et al. (2021) examined how the Rail Operation Control Center formed by Dutch railroad companies resolved disruptions to rail operations and coordinated emergency responses. Interview and observational data indicated that unplanned and emergent activities—informal mutual adjustment—were more influential than planned activities in enacting coordinative communication but were also subject to breakdown when information overload occurred.
Other MTS research indicates limitations in lower-level mutual adjustment's ability to provide effective inter-team coordination. In one study, Davison et al. (2012), conducted a laboratory analysis of integration and component teams and found that reliance on mutual adjustment among component team members was associated with increased performance within component teams but decreased MTS performance, a set of findings that the researchers attributed to breakdowns in inter-team coordination due to the large size and specialized teams. Fodor and Flestea (2016), in a network analysis of a mission to rescue survivors of a failed Romanian medical flight, found that coordination by mutual adjustment alone faltered in instances when new teams entered the MTS and existing communication patterns had to be revised on the fly. Results also indicated that decentralized communication impeded the effectiveness of mutual adjustment among MTS components due to the dissemination of contradictory information unchecked by a central source. Ziegert et al. (2022), in laboratory study, found that a balance of mutual adjustment within and between component teams was necessary to ensure MTS effectiveness. Absent that balance, conflict among component teams increased and MTS performance declined.
In summary, MTS studies have shown mutual adjustment among teams to be an important coordination mechanism in multiteam systems (Hoegl et al., 2004; Nyrud & Stray, 2017), and have indicated that mutual adjustment is an effective inter-team coordination mechanism in systems with manageable information loads (De Vries et al., 2021). Research has also indicated that mutual adjustment loses efficacy in systems characterized by information overload due to larger MTS size (Davison et al., 2012), specialized team-level tasks (Davison et al., 2012), team-level membership change (Fodor & Flestea, 2016), unchecked contradictory information (Fodor & Flestea, 2016) or an imbalance between intra-team and inter-team communication (Ziegert et al., 2022).
Direct supervision
MTS theorists have identified direct supervision as an important inter-team coordination mechanism (e.g., Davison & Hollenbeck, 2012; Guthrie et al., 2005; Luciano et al., 2018; Mathieu et al., 2001, 2018; O’Leary et al., 2012; Rentsch & Staniewicz, 2012; Rico et al., 2018; Zaccaro & DeChurch, 2012). It has been conceptualized in MTS research as leadership (DeChurch, 2002), vertical communication (Hoegl & Weinkauf, 2005), and hierarchy (Davison & Hollenbeck, 2012).
A significant stream of empirical research has examined direct supervision in multiteam systems under the guise of leadership. Early in this stream, DeChurch (2002) and DeChurch and Marks (2006) performed a laboratory study of multiteam systems composed of three teams, one of which functioned as its system's leader team, and found that leader team training in strategy and coordination processes resulted in enhanced levels of inter-team coordination. Subsequently, based on analyses of archival cases of MTS coordination, DeChurch et al. (2011) found that MTS leaders coordinated inter-team efforts by adjusting strategy and coordination processes, i.e., by modifying the roles of component teams as work situations changed, expediting the communication needed to organize team tasks, and managing the flow and timing of the information required to keep all component teams equally informed.
Hoegl and Weinkauf (2005) reported that horizontal communication, i.e., mutual adjustment, proved to be an effective means of inter-team coordination during early phases of multiteam product development projects but that direct supervision in the form of vertical communication became more effective during later phases that required large amounts of task-related information to be distributed in order to coordinate complex work activities. Johannessen et al. (2012) identified a similar information capacity effect, finding that dynamic, complex, hazardous work conditions in offshore maintenance operations required that inter-team relations be coordinated by leadership personnel within a system of centralized hierarchical control. However, a laboratory study by Matusik et al. (2022) indicated that direct supervision (hierarchy) among MTS teams reduced MTS coordination. The researchers attributed this finding to decreased horizontal communication (mutual adjustment) and consequent reduction in the alignment and synchronization of inter-team activities.
De Vries et al. (2016) examined the effects of “integration teams” charged with ensuring effective coordination between the component teams of multiteam systems. Results indicated that inter-team coordination by a hierarchically central team holding integrative leadership responsibilities was successful in enacting effective inter-team coordination. However, Lanaj et al. (2018), in a study of leader and component teams found that the performance of an MTS charged with decision making was highest when a multiteam system's leadership team and component teams held divergent views regarding risk in decision making, and that performance decreased as teams held more convergent views.
In sum, MTS research indicates that direct supervision is an effective means of coordinating inter-team MTS interdependence (De Vries et al., 2016; DeChurch, 2002; DeChurch et al., 2011; DeChurch & Marks, 2006). Research has also indicated that elevated task-related communication requirements (Hoegl & Weinkauf, 2005), dynamic work conditions (Johannessen et al., 2012), decreased mutual adjustment (Matusik et al., 2022), and cognitive convergence regarding decision making tasks (Lanaj et al., 2018) can reduce the ability of direct supervision to coordinate MTS activities.
Standardization
Coordination by standardization has been conceptualized in MTS research as standardization (De Vries et al., 2021) and as planning, including goal setting (e.g., Davison & Hollenbeck, 2012; DeChurch et al., 2011; Dietrich et al., 2013; Luciano et al., 2018; Rico et al., 2017; Scheerer et al., 2014). Shared mental models and comparable ideas—e.g., team mental models, cognitive similarity, shared representations, shared understanding—are also considered a form of standardization in this article. All involve shared mental representations that enable MTS teams to economize on communication and rely instead on common understandings to coordinate inter-team activities. Shared mental models and similar concepts lie at the heart of MTS theory (Asencio & DeChurch, 2017; Guthrie et al., 2005; Hinsz & Betts, 2012; Jimenez-Rodriguez, 2012; Luciano et al., 2018; Mathieu et al., 2001, 2018; Owen et al., 2013; Rentsch & Staniewicz, 2012), but have not been considered a standardization mechanism in prior theory.
Only a single analysis exists of standardization per se in multiteam systems, due to the apparent conflict between the stability needed for standardization to sustain interdependence and the adaptability often required for MTS effectiveness. In that analysis, De Vries et al. (2021), cited previously, reported that inter-team relations in the company of routine MTS tasks were coordinated through standard protocols and prescribed plans supported by a shared information system.
In the single available study of MTS coordination by planning, Lanaj et al. (2013), hypothesized that decentralized planning would increase inter-team coordination failures due to the reduced influence of a common coordinating mechanism. Results of a laboratory experiment confirmed the researchers’ expectations and showed evidence of an overall negative effect on MTS performance.
In MTS research on shared mental models and related concepts, Johannessen et al. (2012), identified shared mental models as an efficient means of coordinating inter-team interdependence. Specifically, the researchers found that shared mental models enhanced coordination by helping teams understand and reinforce standard operating procedures, anticipate setbacks, and minimize the occurrence of operational problems. Murase et al. (2014), in a reanalysis of data from DeChurch and Marks (2006), reported that leaders’ mental models communicated to component teams enhanced inter-team coordination by standardizing the mental models held by members of MTS component teams. Inter-team coordination and MTS performance increased as a result. Lei et al. (2022) reported the results of a large-scale survey that indicated that mental models shared among the leaders of MTS component teams were associated with increased efficiency in new product development due to improvements in MTS boundary management under conditions of environmental uncertainty.
Firth et al. (2015) examined the effects of training designed to reduce perceptual differences among teams and found that it improved MTS performance by enhancing inter-team coordination, suggesting that standardized knowledge is an effective coordination mechanism. However, DeChurch (2002) reported that leader strategy training, in which leaders were trained to communicate a plan of action that specified the order and timing of team actions, resulted in greater accuracy but not in greater similarity in the mental models that indicated how teams should work with each other. This finding suggests that boundary conditions may exist regarding the effectiveness of knowledge standardization via training.
Healey et al. (2009) examined the effects of team transactive memory, consisting of shared systems for encoding, storing, and retrieving information, and predicted that team transactive memory would help responders avoid transmitting unnecessary, ambiguous, or inaccurate information. Findings supported the researchers’ prediction. In a pair of studies, De Vries et al. (2014) hypothesized that cross functional experience would enhance interpersonal cognitive complexity and that this complexity would facilitate inter-team coordination. Results supported hypothesized expectations in instances where team members identified strongly with the MTS. Findings of a second study replicated these results.
To summarize, MTS research on standardization, planning and goal setting, and shared mental models, all conceptualized in this article as variants of the standardization coordination mechanism, indicates that inter-team standardization can enact inter-team coordination (De Vries et al., 2021), in the manner described by Mintzberg (1979, 1993). Planning, a prerequisite to standardization, is effective in establishing inter-team coordination when bifurcation of interests is disallowed (Lanaj et al., 2013). Shared mental models, a specific form of standardization, reduce the need for communication (Healey et al., 2009) and improve MTS boundary management (Lei et al., 2022). They are effective as inter-team coordination mechanisms when assurances exist that they are truly shared among all teams in an MTS. Such assurances can originate in leadership direction, extended experience in the MTS, identification with the MTS, or homogeny training (De Vries et al., 2014; DeChurch, 2002; Firth et al., 2015; Johannessen et al., 2012; Murase et al., 2014).
Higher-Order mutual adjustment
|Mintzberg (1979, 1993) identified a higher-order form of mutual adjustment, distinguishable from lower-order mutual adjustment because it is directed not toward coordination among MTS teams but instead toward the coordination of other inter-team coordination mechanisms already in use. For example, face-to-face communication in MTS leadership meetings might be used to determine which standardized procedure to enact in light of emerging work demands or which team leader to empower to resolve an inter-team conflict. MTS research has examined various mechanisms that introduce higher-order mutual adjustment, including liaison positions and roles, boundary spanning and boundary spanners, and meetings and in particular leadership or command team meetings, all of which are lateral linkages (Galbraith, 1973) that enable the reemergence of mutual adjustment under information-rich, dynamic conditions.
Liaison positions are specialized roles designed to facilitate communication between two or more interdependent teams and to bypass other less direct lines of communication (Galbraith, 1973). Liaisons share coordinative information through face-to-face interaction with the members of bridged teams but lack the formal authority to mandate coordinated action (Galbraith, 1977). Research on MTS liaisons consists of the previously cited field study by De Vries et al. (2021) that examined the actions of a “provisional hierarchy” of liaisons drawn from each of the MTS teams in their study. Mutual adjustment among the liaisons shielded other team members from unwanted distractions and enabled development of a shared understanding of emerging problems that led to coordinated responses. Liaison activity was especially important during the initial coordination of responses to non-routine disruptions, but receded in importance during later phases as standardized procedures were enacted in response to increasing routine.
Boundary spanners are individuals within units that bridge the boundary between their units, whether teams or multiteam systems, and the surrounding context in order to facilitate adjustment to contextual conditions (Thompson, 1967). Boundary spanners use informal communication to forward contextual information to their home unit and are thus a means of introducing mutual adjustment into situations otherwise coordinated by different mechanisms. Davison et al. (2012) found that structured higher-order mutual adjustment enacted by the boundary spanners of component teams most critical to MTS success (point teams) between their component team and leadership teams had positive effects on coordinated MTS performance, but that similar mutual adjustment performed by boundary spanners of less critical (support) teams between their component and leadership teams did not. In a study of aircrew coordination between the cabin and cockpit teams, Bienefeld and Grote (2014) observed simulations of a “smoke in cabin” emergency and examined the effects of shared leadership within and among component teams. Results indicated that coordinative leadership was associated with higher performance of component teams and of the MTS as a whole.
Meetings provide the momentary propinquity required to enable otherwise separated individuals, often acting as team representatives, to communicate and adjust using informal processes. They are a “prime vehicle used in [an] organization to facilitate mutual adjustment” (Mintzberg, 1979, p. 163). In MTS research on higher-order mutual adjustment in meetings, Uitdewilligen and Waller (2012) examined the operations of a temporary multiteam system established by emergency management units and found that non-emergency response was typically coordinated within and among MTS teams using standard procedures, but that emergency response was coordinated by a leader team—the emergency response organization—through the use of higher-order mutual adjustment among the team's members, each of whom was a representative of one of the multiteam system's component teams. Dietrich et al. (2013) reported that meetings between team leaders or within specific coordination teams were the most frequently mentioned means of inter-team under conditions of complexity and uncertainty, and that mutual adjustment within regularly scheduled coordination meetings enabled the multiteam systems to exchange the necessary knowledge required to solve coordination problems and adjust for emerging changes. Finally, in two case studies of inter-team coordination, Dingsoyr et al. (2017) and Dingsoyr et al. (2018) reported that meetings were used extensively as coordination mechanisms in order to deal with task-related uncertainty and change, and that meetings transitioned from scheduled to unscheduled events as task requirements stabilized. The layered mutual adjustment observed in their study was a primary source of inter-team coordination.
In sum, MTS research indicates that higher-order mutual adjustment can be introduced through the incorporation of liaisons, boundary spanners, and meetings, and that such incorporation has beneficial effects on MTS inter-team coordination and performance. In particular, mutual adjustment among liaisons representing their component teams enables the information distribution required during the initial coordination of non-routine disturbances (De Vries et al., 2021). Boundary spanners have the greatest effect on inter-team coordination when representing teams whose actions are the most essential to MTS success (Davison et al., 2012). Their coordinative activities can be so influential that they enhance the performance of teams outside of the immediate boundary spanning relationship (Bienefeld & Grote, 2014). Meetings, and especially meetings within leader teams and among the representatives of interdependent teams, allow the higher-order mutual adjustment to occur that is needed to coordinate inter-team relations in the face of threat, uncertainty, and change. (Dietrich et al., 2013; Dingsoyr et al., 2017; Uitdewilligen & Waller, 2012)
Transitions Among mechanisms: future research
The preceding review, summarized in Table 2, indicates that MTS research has examined the presence and coordinative effects of each of Mintzberg's mechanisms in studies of inter-team coordination. In addition, two studies have produced limited evidence of specific transitions among coordination mechanisms that are consistent with Mintzberg's account. In one study, Hoegl and Weinkauf (2005) verified that direct supervision rises in importance and mutual adjustment recedes in the face of increasing flows of coordinative information. In the other study, De Vries et al. (2021) reported evidence of a transition from standardization to higher-order mutual adjustment as greater coordinative flexibility was needed among MTS teams. Outside of these two instances, MTS research has not yet examined the sequenced ordering of transitions theorized in Mintzberg's model. Understanding this ordering is a necessary prerequisite to efforts to develop predictive models useful in planning coordinative strategies and enacting successful inter-team coordination.
Coordination mechanism types and research results.
Sequence of transitions
Mutual adjustment is the simplest of the three coordination mechanisms identified by Mintzberg, requiring little more than a common language and interpersonal contact, and succeeds when the information needed to coordinate structural interdependence is limited in amount. As a means of inter-team coordination, effective mutual adjustment requires that every interdependent team be able to communicate with all other interdependent teams in a completely connected network. If a team is left out of the communication network, that team will not receive the information needed to coordinate effectively. Due to the requirement of universal involvement, only a limited number of teams can be effectively coordinated by mutual adjustment alone.
In instances where greater information requirements undermine the ability of mutual adjustment to distribute coordinative information, direct supervision emerges as an effective alternative (e.g., Miller, 1959). Direct supervision requires that hierarchy be established, so is more difficult than mutual adjustment to initiate and sustain. However, having one individual serve as a central information node and direct the work of interdependent teams reduces the number of communication linkages required to disseminate coordinative information and thus overcomes the information-distribution limitations of mutual adjustment (Caplow, 1964; Davison & Hollenbeck, 2012). Since direct supervision reduces the number of linkages used to distribute information, it also increases the speed of information distribution. However, direct supervision has its limits, since even greater information requirements will overload the supervisor in the role of centralized information node. Given information overload, direct supervision falters and coordination declines if it is not sustained by another means.
In the face of coordinative information overload, standardization is able to reduce the amount of information that must be distributed to coordinate effectively and for this reason becomes attractive as a means of sustaining inter-team interdependence. Following logic developed by March and Simon (1958), if the coordination of interdependent teams is incorporated into the standards established to guide performance, coordination can be achieved by following standard procedures without additional communication. Information exchange that would otherwise overburden mutual adjustment or overload direct supervision is lessened if not completely eliminated.
Standardization requires planning and implementation procedures that make it more difficult to effect than either mutual adjustment or direct supervision. In addition, its efficacy is limited because change renders plans invalid and standard procedures unable to coordinate. Coordination by standardization requires a level of stability that permits the recurrent use of minimally altered procedures (Brunsson et al., 2012; March & Simon, 1958). Where standardization has been rendered ineffective by change, a higher-order form of mutual adjustment emerges as the predominant means of distributing coordinative information. As previously described, its function amounts to the “coordination of coordination.”
To complete his model of basic coordination mechanisms, Mintzberg indicated that mechanisms surpassed in the transition from mutual adjustment to direct supervision to standardization to mutual adjustment do not disappear but instead continue to function as secondary means of coordination. Thus, in situations where coordinative information has overloaded mutual adjustment and necessitated coordination by direct supervision, mutual adjustment remains in place as a secondary mechanism that fills in gaps in the information distributed by the supervisor (as noted by Matusik et al., 2022). Where direct supervision has been replaced by standardization, direct supervision, and mutual adjustment continue as supplementary sources of coordinative information when plans are incomplete. Finally, where standardization has been replaced by mutual adjustment, three alternatives—standardization, direct supervision, and lower-order mutual adjustment—remain active as secondary sources of coordination.
Given that little is currently known about the transitions that occur among the basic coordination mechanisms active in multiteam systems and that these transitions are essential parts of MTS coordinative processes, it appears reasonable to suggest that future MTS research should be directed toward investigating the ordering and causal logic of the transitions described in Mintzberg's model. Beyond this research, the simultaneous presence of multiple coordination mechanisms invites examination of the distribution of coordinative processes between mechanisms when two or more are simultaneously active. It is not known, for example, whether lower-order mutual adjustment's decay occurs in the form of an immediate step function or instead recedes in a gradual slope, nor is it known exactly how secondary lower-order mutual adjustment supplements primary direct supervision. Similarly, the combined relationship between standardization and secondary direct supervision and lower-level mutual adjustment is unexplored. Finally, the degree of change required to produce a decline in standardization's coordinative efficacy and lead to the reemergence of mutual adjustment, now in a higher-order form, has yet to be investigated.
Precipitant factors
In failing to direct significant attention to the existence and consequences of Mintzberg's sequence of transitions, MTS research has also failed to examine the variety of factors that can stimulate and shape this sequence. The value of research on these precipitants, which would extend beyond the conceptual boundaries of Mintzberg's basic model, lies in the development of dynamic modeling able to explain how and why transitions occur among inter-team coordination mechanisms.
As theorized by Mintzberg (1979, 1993) and other researchers (e.g., Daft & Lengel, 1984; Galbraith, 1977), lower-order mutual adjustment is limited in the amount of information it can distribute and is susceptible to information overload. MTS research provides evidence of this effect (Davison et al., 2012; De Vries et al., 2021). Among the teams of an MTS, such overload can originate in any of three sources. First, the number of interdependent teams to be coordinated influences the ability of mutual adjustment alone to provide necessary coordination. As the number of linked teams increases, the number of linkages required to communicate coordinative information in the completely connected network required to keep all interdependent teams fully informed rises at an accelerating rate (Bossard, 1945; Caplow, 1964; Davison & Hollenbeck, 2012). To illustrate, a completely connected network of three teams requires three links, a completely connected network of five teams requires 10 links, and a network of ten teams requires 45 links. Information overload occurs as communication requirements rise due to the increasing number of linkages.
Second, the density of the task-related interdependence (as distinct from the communication linkages just described) shared among parties also affects lower-order mutual adjustment's coordinative efficacy. Typical models of interdependence identify four specific types (e.g., Saavedra et al., 1993; Shuffler et al., 2015). In pooled interdependence, the least dense form of interdependence, the teams in an MTS share dependence on a common resource but are not directly dependent on one another. In sequential interdependence, teams share one-way relationships with one another as would be exemplified by an assembly-line type of inter-team arrangement. In reciprocal interdependence, relationships among teams are two-way, but not all teams share interdependence with one another. In intensive interdependence, the densest form of interdependence, every team shares a two-way linkage with every other team in the MTS. As indicated by Thompson (1967), as increases occur in interdependence density, from pooled to sequential to reciprocal to intensive, the information that must be exchanged to monitor the robustness and continuity of each instance of interdependence also increases. This additional information contributes to eventual overload that threatens to ability to communicate coordination requirements through face-to-face exchange.
Third, the complexity of the tasks performed by interdependent teams influences whether lower-order mutual adjustment alone can enact and sustain interdependence. Complex tasks require the exchange of information to understand task requirements and perform task obligations, and greater task complexity increases the task-related information that must be shared among interdependent performers (Hærem et al., 2015; March & Simon, 1958; Wood, 1986). At some point, task-related information requirements, in combination with coordinative information exchange, overwhelm the ability of interdependent teams to exchange information and manage interdependence among themselves.
Direct supervision, by substituting a single communication hub for the completely connected network of mutual adjustment, substantially reduces the number of communication linkages necessary to inform all teams and coordinate inter-team interdependence. This shift is seen in MTS research reported by Hoegl and Weinkauf (2005) and Johannessen et al. (2012). However, the single hub is subject to overload as coordinative information continues to increase. Consequently, just as the number of linkages required to distribute information, the density of interdependence, and the level of task complexity affect the coordinative efficacy of mutual adjustment, so to do they influence the effectiveness of direct supervision. Increases in any or all of these precipitants can eventually overload the central hub and undermine direct supervision's ability to provide necessary coordination. Standardization emerges as the primary coordination mechanism, as observed in MTS research by Healey et al. (2009) and De Vries et al. (2021).
Limitations to standardization's ability to provide necessary coordination also involve information considerations, now in the form of increasing quantity or modulations that were unanticipated during the design of standardized procedures. For example, medical units that have trained together to develop standardized procedures to synchronize inter-team efforts will find it necessary to resort to other means of coordination if additional units are required to join their efforts in order to respond to unplanned conditions. Those additional units bring with them questions about how to integrate and coordinate efforts that are not envisioned in the original standard procedures. Mutual adjustment reemerges as the primary means of coordinating inter-team relationships, albeit now of the higher-order type. In MTS research, this reemergence was documented by Dietrich et al. (2013). Within this type of coordination, leaders or leadership units acquire coordinative information through liaisons, boundary spanners, or meetings, and reshape inter-team coordination procedures accordingly.
As indicated by this theoretical description, the amount and type of coordinative information required to harmonize inter-team efforts is a primary precipitant that stimulates transitions among mutual adjustment, direct supervision, and standardization, and is offered as an important topic for future research in inter-team coordination. Information-related overload effects have been identified as reducing the efficacy of specific coordination mechanisms in existing research (e.g., Davison et al., 2012; Dietrich et al., 2013; Johannessen et al., 2012) but have been recognized as transition precipitants in only two studies (De Vries et al., 2021; Hoegl & Weinkauf, 2005). Within research on information effects, MTS inter-team analyses could seek answers to questions concerning boundary ranges in the number of linkages, density of interdependence, level of task complexity, or combination contributing to information overload in the case of lower-order mutual adjustment and direct supervision, or unanticipated information requirements in instances of coordination by standardization. In addition, MTS studies could examine how the number of linkages, density of interdependence, and level of task complexity interact to exert combined effects.
Additional research
Precipitants that lie beyond Mintzberg's model, many of which have been deemed important for other reasons in research on organizations and organizational behavior, are also suggested as areas for future research on inter-team coordination. Cognitive diversity is an example of this type of precipitant. It originates in the different life experiences of people from different races, genders, education, functional specializations, etc., and increases the need for information exchange in order to develop shared understandings and enact successful coordination (e.g., Larkey, 1996; Roberson, 2006; Thomas & Ely, 1996). Diversity within and among teams should be expected to add to the information exchange required to coordinate inter-team interdependence Consequently, it is another source of potential overload. Paradoxically, speculation suggests that information overload attributable in part to diversity-related differences can lead to standardization, which in turn will eliminate the information exchange and gains in understanding that are significant benefits of diversity in organizations.
Inter-team analyses should also examine possible boundary conditions that limit the applicability of the model reviewed and further developed in this article. For instance, research on teams has differentiated between functional teams consisting of members who perform the same organizational function and combine with other functional teams to create complete workflows, and divisional teams made up of members who perform different functions and span entire workflows within their boundaries (e.g., Hollenbeck et al., 2002). Implications with respect to inter-team coordination are clear, since divisional teams can function independently whereas functional teams must coordinate among themselves. Galbraith (1973, 1977) described the formation of self-contained groups, i.e., groups able to complete entire work sequences using their own resources and independent of other groups, as a way to minimize coordinative requirements among groups. Similar logic suggests that the functional versus divisional distinction between multiteam systems should have a significant effect on inter-team coordination requirements and on the progression of mechanisms examined in the sequence described by Mintzberg (1979, 1993).
Finally, Mintzberg's model and the research and theory that it incorporated were published before the advent of contemporary information processing capabilities. Information technology now available can supplement the face-to-face exchange of mutual adjustment as classically defined, enabling otherwise overloaded processes of information exchange and dissemination to retain efficacy (e.g., Andersen, 2001; Shen & Shaw, 2004; Sor, 2004). For example, “respond to all” email messages can quickly and efficiently distribute shared information that would otherwise clog the communication channels of mutual adjustment processes. Information preparation and distribution that formerly required minutes and multiple interactions can now be accomplished in seconds through a single technology-supported exchange.
The potential effect of this difference on mutual adjustments’ coordinative capability appears to be substantial, possibly extending the scope of mutual adjustment's effectiveness well beyond the information-carrying limits of face-to-face interaction. The probable result is a delay in the transition to direct supervision and a flat MTS structure instead of hierarchy. Concepts often examined in MTS research, repositioned as moderators, are likely to influence this process. For example, the possession of shared mental models is likely to facilitate the mutual interpretation of inter-team email messages as intended, whereas the absence of this mental commonality seems likely to increase the incidence of differing interpretations among the teams, each with its own goals and aspirations, that comprise a multiteam system.
In addition, the same information technology appears likely to extend the effectiveness of direct supervision by assisting the supervisor in formulating and distributing coordinative information, resulting in an increase in the number of subordinates that a direct supervisor can effectively coordinate (Melin & Axelsson, 2005). Shared supervisory messages can be communicated using emails or texts, leaving only individualized messages to be communicated in one-on-one interactions. Likely to be affected are the ranges in the number of linkages, density of interdependence, and levels of task complexity that would be identified as transition points in the research on direct supervision in multiteam systems.
MTS research has developed a noteworthy theoretical base regarding the importance of information technology and its effects on inter-team coordination. For example, DeChurch and Zaccaro (2010) indicated that technology-mediated communication is important due to the complexity and scope of the tasks typically performed by multiteam systems, and also noted the importance of information technology in supporting communications among MTS teams that are geographically dispersed. Citing studies by Choi et al. (2010) and Wu and Chen (2012), Luciano et al. (2018) suggested that information technology systems enable multiteam systems to perform complex tasks such as managing congested airspace or coordinating firefighters in a burning structure. Similarly, Dean (2021) cited a study of offshore petroleum production by Larsen et al. (2014) to exemplify the use of information technology to supplement cross-functional communication. In all cases, existing theory and research support the idea that information technology has the potential to increase the efficacy of each of the MTS coordination mechanisms examined in this article.
Conclusion
Inter-team coordination is a central focus in research on multiteam systems and has been so since initial MTS conceptualization by Mathieu et al. (2001). A significant stream of MTS research on inter-team coordination has assessed the presence and efficacy of four basic coordination mechanisms—lower-order mutual adjustment, direct supervision, standardization, and higher-order mutual adjustment. Yet to be examined in MTS research are the complete configuration of transitions among the coordination mechanisms as well as the effects of precipitants that stimulate these transitions. Research in these areas would contribute novel additions to knowledge about MTS coordination and, in the process, would lead to the development of a more compelling model of inter-team coordination in multiteam systems. The theoretical and practical significance of these contributions support their future pursuit.
Footnotes
Funding
The author received no financial support for the research, authorship, and/or publication of this article
