Introduction
Teams have become a go-to tool for organizations attempting to create value in the modern world. Researchers and practitioners alike have addressed this trend by trying to help organizations understand the determinants of team success with the end goal of helping organizations to design and cultivate productive teams. While this work has been valuable, in this paper, we introduce the counterintuitive idea that there are risks associated with high team performance. That is, to the extent to which a team achieves a certain level of success, we explore whether there are potential pitfalls that may become more likely precisely because of that success.
A lot of academic energy has been spent developing models of team performance (Campion et al., Reference Campion, Medsker and Higgs1993) with many of these models examining how team inputs, team processes and team emergent states shape overall organizational team performance (Mathieu et al., Reference Mathieu, Maynard, Rapp and Gilson2008). The implicit goal of these models is to help us understand what factors influence how well teams’ function and accomplish their goals. Here, we will consider five categories of risk-factors or pitfalls that may be more likely when a team is successful. If the risks are acknowledged and managed properly a team is more likely to survive its own success. What are the potential downsides of team success?
Potential Downsides of High Performance
Groupthink: All for One and One for All?
Groupthink has a 50-year history in our academic literatures and represents an ineffective group decision-making process in which maintenance of group cohesiveness obstructs the team’s critical thinking (Janis, Reference Janis1971, Reference Janis2008). The success of a team could serve to increase several risk factors of groupthink. An Illusion of Invulnerability, a key symptom of groupthink could be made more likely by the team’s past success. A second symptom, a Belief in the Inherent Morality of the group, might also be increased by the group’s past success. This assumed morality would grow out of cognitive heuristics and cognitive dissonance that suggest that when an individual or team is successful on a given task, then the team must have been just and good. Finally, just as positive distributive justice can minimize critical focus on procedural justice (Colquitt et al., Reference Colquitt, Noe and Jackson2002; Mumford, Reference Mumford2010), the success of the team would likely make self-censorship and its accompanying Illusions of Unanimity more likely.
Overconfidence: Pride Goeth Before the Fall?
The construct of overconfidence has a rich tradition of being studied as a cognitive bias (Kahneman & Tversky, Reference Kahneman and Tversky1977) and as a leadership failing (Shipman & Mumford, Reference Shipman and Mumford2011), while also being considered as a leadership asset in creating innovation (Galasso & Simcoe, Reference Galasso and Simcoe2011), but it has only more recently received direct attention in the teams literature. For example, it has been found to be a contributing factor in poor team decision-making (Meissner et al., Reference Meissner, Schubert and Wulf2018), while also being associated with improvements in team performance (Gervais & Goldstein, Reference Gervais and Goldstein2005).
So, when a team experiences success in the short-term this is likely to increase team confidence and potency and by so doing also increase the risk of overconfidence bias. While teams that are successful may translate into the potential for continued high performance in the future (Gervais & Goldstein, Reference Gervais and Goldstein2005), to the extent to which the team’s performance context changes and strategic decisions are needed, the results may be different. For example, the success of the team may result in a lack of “bad news” needed to help the team ‘snap out of’ its collective blindness to risks of less favorable futures (Bénabou, Reference Bénabou2013). In addition, consistent with the halo effect and self-serving bias, successful teams may be more likely to attribute their success to themselves and their leadership as opposed to situational factors. This would make it less likely that they are able to detect deficiencies in their team and have inflated expectations of future successes (Shipman & Mumford, Reference Shipman and Mumford2011).
Regression to the Mean: Back to Normal?
Regression to the mean is an established concept that illustrates that extreme examples of positive or negative performance are likely to shift closer to average during future performance episodes (Barnett et al., Reference Barnett, van der Pols and Dobson2005; Bland & Altman, Reference Bland and Altman1994; Cook et al., Reference Cook, Campbell and Day1979).
Because virtually all episodic performance measurement includes random variation, extreme high team performance is likely to be attributable, at least in part, to that random variance. That suggests that when a team experiences success they may be more likely to experience lower future performance due to the random variability determinants of performance. While regression to the mean has received attention in longitudinal studies guarding against misattributing performance improvement to an intervention (Cannella Jr & Rowe, Reference Cannella and Rowe1995); in the case of successful teams, regression to the mean would only be manifest in subsequent performance episodes. This suggests that successful teams should be extra vigilant of the risks of performance declines, a vigilance that factors such as groupthink and overconfidence may make less likely.
Roles: Too Much of a Good Thing?
Many decades ago, when team structures were gaining popularity, it may have been likely for employees to work on single teams. However, in many contemporary organizations teams have become the default organizing structure (Devine et al., Reference Devine, Clayton, Philips, Dunford and Melner1999). As a result of such cross-functional teams in matrix organization designs (Holland et al., Reference Holland, Gaston and Gomes2000) and multi-team systems (Mathieu et al., Reference Mathieu, Marks and Zaccaro2002), it is probable that employees are members of multiple distinct teams (Maynard et al., Reference Maynard, Mathieu, Rapp and Gilson2012; O'leary et al., Reference O'leary, Mortensen and Woolley2011). While there may be some synergy, overlap and positive consequences that may emerge from such multiple team membership, there are at least three distinct problems that may emerge from multiple team membership in successful teams in this environment.
First, team success can lead to role conflict (Beauchamp & Bray, Reference Beauchamp and Bray2001; Ebbers & Wijnberg, Reference Ebbers and Wijnberg2017). Employees in team-based organizations are forced to manage their time, effort, and priority allocations across these multiple teams. As a result, if one team is more successful than another, it may be that employees will choose to invest more of their resources into the team experiencing success and engage in more social-loafing oriented approaches to the lower-performing team. A lack of balance may be made more likely by differential success across the teams.
Second, team success may lead to role overload. Within teams with the organic organization of self-managed teams, it is likely that employees on high-performing teams are asked to be on more teams. Thus, a consequence of successful teams may be that their comprising members experience role overload as the number of teams they are involved in grows and their roles within these teams is augmented. At the individual level this has been called job engorgement (Campion et al., Reference Campion, Mumford, Morgeson and Nahrgang2005) but it needs specific research within the teams literature.
Third, team success may lead to team members being poached by other teams, as is often seen in sports teams (Langelett, Reference Langelett2003). When a team achieves high performance, it is more likely that those team members are seen as being high-value employees both internal and external to the organization. Internally, if company policies limit the size of the employee’s role to avoid the role overload issue, the result of the team success might be that the team loses the employee to another team. Externally, the team’s success may lead to other organizations hiring the employee away from the team. In either case, the high performing team would now be experiencing a loss in human capital which could have detrimental consequences for future performance.
Task Strategy Calcification: If It Ain’t Broke, Don’t Fix It?
Teams adopt norms for how they will interact with each other and approach the work (Birenbaum & Sagarin, Reference Birenbaum and Sagarin1976). The presence of such norms has traditionally been used as one of the defining features of a team (Gammage et al., Reference Gammage, Carron and Estabrooks2001; Hackman, Reference Hackman and Lorsch1987) and have been thought to have valuable outcomes for the team. Specifically, norms have been suggested to help team members know what to expect of each other and to integrate their actions over time and has been studied as entrainment (Harrison et al., Reference Harrison, Mohammed, McGrath, Florey and Vanderstoep2003). In addition, team members form identities wherein their sense of self and the team are closely connected (Hinds & Mortensen, Reference Hinds and Mortensen2005). With shared identities, the team’s success is experienced as the individual’s success and a risk to the team may be felt as a risk for the individual.
As a result, successful teams may experience a calcification of task strategies for at least two reasons. First, because the team has become proficient approaching its work in a given way, it will have established entrained patterns of individual and team-related behavior. These patterns may be functional in a static task environment. However, within a more dynamic task environment, the teams’ past strategies may not be well suited to the task and may result in suboptimal future performance.
Second, the successful team will face increased risk associated with task strategy switching. With the success, the team has created internal identities and external reputations for high performance. With these heightened internal and external reputations, there would likely be a greater sense of risk associated with trying new task strategies. Again, in a dynamic task environment where innovative task strategies are needed, this can prove problematic for future performance. Likewise, with these internal identities created, if subsequent performance is hindered, it has an even greater impact on the individuals of such teams.
Implications
Recommendations for Teams Research
Given the research and ideas reviewed here, there are several important implications for moving our research literatures forward. First, longitudinal research design matters. Most of the assertions made here are empirical questions subject to empirical examination. However, their validity can only be fully investigated with longitudinal research designs that capture team performance over a longer-term time horizon. For example, any regression to the mean effect or role overload effect would not be manifest in the short term. Only by considering longer term performance lagged from the initial performance episode could such effects be measured and evaluated.
Second, these arguments have implications for the criteria problem within the team’s literature. The extent to which these issues are valid and problematic can only be discovered if we design studies with criteria that are broad enough to capture not just past performance but also current challenges and anticipated future performance. Consider the value of routinely informing our studies with forward-leaning criteria such as team burnout, anticipated membership changes, aspirational future performance and the like. It may be that by including future-oriented constructs and measures we can more fully uncover performance-related processes that are masked by unidimensional performance measures. Ensuring that our theoretical frameworks defining team effectiveness to include viability is a step in the right direction, but those constructs have been slow to become standard practice for assessing our team models (Bell & Marentette, Reference Bell and Marentette2011).
Finally, the arguments laid out here call for more integrative research into our models of team effectiveness. Practical considerations often limit the scope of studies and the evolution of literatures often lead to specific and narrow studies. These arguments suggest that there is value in considering a greater breadth and depth of ideas. For example, while overconfidence bias is studied at the individual level and in the leadership literatures, its corollary concept of groupthink has received less attention. A deeper investigation into these ideas while connecting them to broader research models will help us provide more valid long-term advice for team effectiveness.
Recommendations for Teams in Practice
In addition to how our recommendations may shape future research on teams, there are also several practical recommendations that flow from these ideas. First, managers must take a realistic and multi-faceted view of the teams they manage. There are both positive and negative outcomes associated with teams and managers should be aware of both. This more complex and nuanced way of viewing teams can help managers have an open mind to avoid potential pitfalls.
Second, these ideas highlight that vigilance may be called for in managing teams, in particular when things are going well. We have called attention to several specific areas for leaders to both monitor and to which they can alert their self-managing teams.
Finally, these issues make it clear that leaders of work teams should adopt a more process-oriented perspective for managing their teams. Through the systematic and continual focus on the processes, positive results will follow. The process focus takes some of the focus off the results that may have variability due to the factors delineated here.
Conclusions
Collaborative work structures such as teams have served organizations well for many decades. This is due, in part, to researchers and leaders who push themselves to better understand the nature and dynamics of teams and how they are implemented in practice. Our hope in providing these arguments regarding the potential pitfalls of high team performance is that these ideas will help both researchers and practitioners continue to refine their understanding and agility in using teams to achieve organizational goals for many decades to come.
