Are Closer Workplace Relationships Always Better?

2026/11/7 6 min read

Background

Good relationships at work make it easier to get things done. When colleagues trust and help one another, communication, information sharing, and coordination tend to run more smoothly. Yet close relationships can also encourage people to spend so much time advising and supporting colleagues that their own work falls behind. This paper develops a theoretical explanation of why workplace relationships that are too weak or too close can both reduce team output.

The central trade-off is simple: helping a colleague advances their work and builds a relationship, but the time spent doing so comes at the expense of one’s own tasks.

Workplace relationships have both professional and personal value. They can make people more productive, while also providing friendship, a sense of belonging, and enjoyment in their own right. But the degree of closeness that members themselves prefer need not be the one that maximizes team output. What does this imply for rewards, team size, and the role of leadership?

What the paper does

We begin with a two-person team. Each member divides a fixed amount of working time between their own tasks and helping the other person, through advice, practical assistance, or similar activities. Helping directly improves the colleague’s task output while also strengthening the relationship between them.

A workplace relationship generates two kinds of benefits. Its productive value comes from making coordination and information sharing easier, thereby improving both members’ productivity. Its private value comes from friendship, belonging, and other benefits that members enjoy independently of work output. We also allow relationship building to be complementary: helping strengthens the relationship more when the colleague is also helping in return.

The starting point is individual performance pay, under which each member is rewarded for their own task output rather than the output their help creates for someone else. We examine when this leads to too little engagement in relationship building, or to so much engagement that individual tasks are neglected. We then study combinations of individual and team rewards, team size, leadership through prior commitment, and differences in members’ preferences.

What it finds

First, there is a level of private value placed on workplace relationships that maximizes team output.

Under individual performance pay, the direct benefit of helping accrues to the colleague. Members who attach little private value to their relationship therefore tend to invest too little time in building it. At the other extreme, members who place very high value on the relationship devote too much time to each other and neglect their own tasks. Neither insufficient nor excessive relationship building maximizes team output. The best outcome lies between these extremes.

Second, leadership can improve performance when productive relationships are underdeveloped, but it reduces performance when relationship building is already sufficient or excessive.

In our model, a leader commits to an allocation of time before the follower chooses theirs. By taking the initiative in helping, the leader encourages the follower to help more as well, strengthening their relationship. This can improve team output when too little time was previously devoted to building productive ties. But where relationship building already takes too much time away from individual tasks, the same leadership mechanism makes performance worse. Remarkably, both the leader and the follower gain in utility even when team output falls. An organizational structure that members welcome is not necessarily the one that produces the most.

Third, small teams may become too absorbed in their interpersonal relationships, while large teams may fail to develop relationships that would help them work effectively.

In a small team, mutual helping reinforces the relationship, which in turn encourages still more engagement. Members may end up spending too much time with one another. Expanding the team spreads each person’s helping across more colleagues and weakens this feedback. That can curb excessive engagement, but a team that grows too large may build too little of the relationships needed for productive work. Under certain conditions, an output-maximizing team size exists, and that size is larger when members place greater private value on their relationships. Team size also affects the productive benefits of relationship formation, so the optimum is not determined by engagement levels alone.

Fourth, making workplace relationships more productive can sometimes reduce team output.

If members kept their time allocations unchanged, a more productive relationship would raise output. But it also makes relationship building more attractive, prompting both members to spend more time helping. When their responses reinforce one another, the loss of time available for individual tasks can outweigh the direct productivity gain. This paradox can persist even after the organization chooses the best available mix of individual and team rewards.

Fifth, the best reward system and team size depend on members’ abilities and how much they value workplace relationships.

Team-based pay gives members a share of the output their help creates for colleagues, encouraging them to help more. For highly skilled members, time taken away from their own tasks has a higher opportunity cost, so individual performance pay alone can discourage helping. Teams with greater member ability may therefore benefit from placing more weight on team performance. Conversely, teams whose members highly value their relationships need less team-based pay, because the members already have a personal incentive to help. And when an interior optimal team size exists under individual rewards, a higher private value of relationships calls for a larger team.

Taken together, the goal is not simply to make workplace relationships stronger, but to achieve an appropriate balance between building those relationships and doing one’s own work.

Why it matters

Policies that encourage communication and helping can have very different effects depending on whether a team has too little or too much engagement in relationship building to begin with. A leader who helps first can strengthen workplace ties, but making an already close team even closer does not necessarily improve its performance.

Team formation and performance evaluation likewise need to account for workplace relationships. Enlarging a small team may help when members spend too much time on one another, but may be counterproductive when productive relationships are already weak. The appropriate balance of individual and team rewards also depends on ability and relationship preferences. Moreover, even among equally skilled colleagues, the person who places more value on the relationship may provide more help while recording lower output on their own tasks. Individual performance measures can therefore miss important contributions to the team.

This is a theoretical study that identifies the conditions under which these outcomes arise. Workplace relationships can improve productivity, but forming them takes time—and that time must come from work that members could otherwise have done on their own tasks.

This research was supported by JSPS KAKENHI Grant Numbers 21K13299 and 24K04912.


About the paper

  • Title: Helping as Relationship Building: Organizational Design and Team Performance
  • Authors: Keisuke Hattori and Mai Yamada
  • Year: 2026
  • Publication status: Working paper, available on EconStor
  • Full paper: https://hdl.handle.net/10419/344455