Verified by 
In many operational settings, like in-flight service, police patrols, and ambulance services, teams are fluid. They are assembled for short collaborations and then disbanded. In these settings, is it better for team members to have familiarity with each other or to be exposed to multiple partners?
We know from prior studies that there are plenty of benefits of familiarity when it comes to teams. They have improved coordination and shared knowledge about the task at hand. However, we don’t know how these benefits compare with those of being exposed to multiple partners over time. We suspect that exposure to multiple partners would enhance creativity and problem-solving, which would translate to improved operational performance, but it has yet to be measured.
In a recent study, my colleagues and I sought to evaluate the performance of both types of teams. We used data from 2011 on ambulance transports from the London Ambulance Services involving new paramedic recruits. Ambulance transports are typically staffed by stable teams of two paramedics. However, new recruits are scheduled on a relief roster and therefore assigned to partners based on administrative convenience. This results in exogenous variation in partner assignments, by which some new recruits have stable long-term partnerships and others get exposed to multiple partners.
This scheduling created an ideal experiential setting for us to examine the performance of both types of teams. Analyzing the data, we jointly estimated the impact of team familiarity and multiple partner exposure on operational performance. We found that the impact of working with multiple partners depended on the type of process at hand.
For less standardized tasks, teams whose new recruits had high multiple partner exposure significantly outperformed teams with more team familiarity. These types of tasks include picking up a patient at the scene where paramedics don’t know what to expect and have to use creativity, deal with family members, and make the area safe. For example, during one of my observational shifts, I saw two female paramedics decide on the spot how best to carry a large man down four flights of stairs. That isn’t a situation likely described in a manual.
For these less standardized tasks, we saw a strong and immediate positive effect for teams comprising paramedics with higher partner exposure. Specifically, we saw that a two standard deviation increase in the multiple partner exposure of a new recruit results in his crew spending on average 7.3 percent less time (2.2 minutes relative to an average of 30.5 minutes) for the task. This is likely to have a significant impact on patient outcomes for approximately 20 percent of the transports in our data that were classified as life-threatening.
However, for the more standardized task of handing the patient over at the hospital, the effect of multiple partner exposure on performance is moderated by cumulative experience. Greater partner exposure is beneficial only once the total number of transports by the new recruit is higher than a threshold (approximately four months on the job). Beyond this level of sufficient experience, a two standard deviation increase in prior partner exposure reduces handover time by 9.8 percent.
The threshold result could be attributed to the fact that in completing a standardized task, workers can rely on standard operating procedures, which guarantee a certain level of performance. Only with sufficient experience do they feel comfortable overwriting the standard operating procedures to enhance their performance.
Both of these types of effects were amplified during periods of high workloads.
Our study suggests that for non-standardized tasks – ones that don’t come with a manual – managers should not blindly favor or adhere to a high-team familiarity strategy because there is value in exposing workers to multiple partners. In other words, for non-standardized tasks there is value in observing a variety of techniques, employed by experienced workers, to accomplish a given task.
For standardized tasks, where workers can rely on operating procedures as a guide, there is still value for exposure to multiple partners. They can help each other develop the skills to outperform the procedural guide.
Overall, inducing high team familiarity by keeping team membership intact can limit workers’ opportunities to acquire useful knowledge and alternative practices from exposure to a broader set of partners. Our analysis shows that managers should consider scheduling strategies that favor exposure to multiple partners over ones based on team familiarity.
Prof. Jónas Jónasson is a professor at the MIT Sloan School of Management and co-author of “Learning for many: Partner exposure and team familiarity in fluid teams,” published by INFORMS.
