Reinvested Effort, Not Persistent Repair or Withdrawal: Stepping Back from a Difficult Working Relationship and Employee Exhaustion and Engagement

Abstract

We followed 418 employees in twelve organizations for six months after each named one working relationship they had tried and failed to make workable. Employees who cut their contact with that colleague back to what shared tasks required and put the freed time into at least two new recurring working contacts elsewhere ("reinvestment," n=141) reported lower emotional exhaustion at six months (3.4/7) than employees who kept working at the difficult relationship ("persistent-repair," n=152; 4.3/7, a 26% gap) or who pulled back while adding fewer than two new recurring contacts ("withdrawal," n=125; 4.1/7, a 21% gap). Their work engagement was higher (5.1/7, versus 4.2/7 and 4.1/7; gaps of 21% and 24%), and managers blind to group rated their collaboration with the wider team higher (5.0/7, versus 4.3/7 and 4.1/7). The path was not smooth: reinvestment employees' exhaustion rose from 4.3/7 at baseline to 4.5/7 at two months, while they spent a logged 2.3 hours a week on their new contacts, before falling. Among 97 distanced colleagues who took part, those on the other side of reinvestment rated their treatment as about as civil as those on the other side of persistent repair (5.0/7 versus 5.2/7), and more civil than those on the other side of withdrawal (4.2/7, a 19% gap).

1. Introduction

Most workplaces carry an unspoken rule that a professional gets along with everyone. When one relationship refuses to work, the rule points in a single direction: try harder. Some people try for years.

Our two units have come at this question from different sides. One has studied volunteers who declined extra requests once they reached their capacity, and found them more likely to still be volunteering six months later; the other has studied how the way people treat their own mistakes shapes how they judge colleagues'. Together they suggested that how people spend limited effort, and how they treat themselves when it falls short, shapes how they treat others. Effort that goes into a relationship with no prospect of improving is also effort that is not going anywhere else.

That leaves open where the effort should go. Pulling back alone may protect a person while leaving them isolated, and the colleague they pulled back from may read the distance as a snub. We compared three responses to a difficult working relationship, measuring exhaustion and engagement in the person who chose, and civility as experienced by the colleague on the other side.

2. Method (Summary)

Employees in twelve organizations (finance, healthcare administration, engineering, and retail head offices) were screened for a working relationship they described as persistently difficult despite at least three months of effort on their part. Of 463 who qualified and enrolled, 418 fit one of the three patterns described below and were followed for six months. Each completed monthly logs of their contact with the named colleague and of any new recurring working contacts, which were checked against calendar metadata (meeting counts, durations, and attendees only; no content).

Classification used the first two months of logs. Employees were "reinvestment" if they reduced discretionary contact with the named colleague to what shared tasks required and established at least two new recurring working contacts (meeting at least every two weeks) with colleagues outside their usual circle (n=141). They were "withdrawal" if they made the same reduction without adding two or more new contacts (n=125), and "persistent-repair" if they kept or increased contact aimed at improving the relationship (n=152). The 45 enrolled employees whose first two months fit none of these patterns, such as a partial reduction in contact, were excluded.

Emotional exhaustion and work engagement were self-reported (1–7) at baseline, two months, and six months. At six months, each employee's line manager, blind to group, rated the employee's collaboration with the wider team (1–7). Every named colleague was invited separately, without being told why they had been named, to rate how civilly they were treated by the enrolled employee (1–7); 97 agreed (33 reinvestment, 36 persistent-repair, 28 withdrawal).

3. Findings

At baseline, exhaustion was similar across groups (4.3/7 reinvestment, 4.2/7 persistent-repair, 4.3/7 withdrawal). At six months it stood at 3.4/7 for reinvestment, 4.3/7 for persistent-repair (a 26% gap), and 4.1/7 for withdrawal (a 21% gap). Persistent-repair and withdrawal did not differ meaningfully from each other.

The six-month figure hides a detour. At two months, reinvestment employees reported 4.5/7 exhaustion, higher than their own baseline and higher than either other group (4.3/7 and 4.0/7). Their calendars showed 2.3 hours a week spent with new contacts over the first eight weeks.

Work engagement at six months was 5.1/7 for reinvestment, 4.2/7 for persistent-repair (a 21% gap), and 4.1/7 for withdrawal (a 24% gap). Manager ratings of collaboration with the wider team followed the same order: 5.0/7, 4.3/7 (a 16% gap), and 4.1/7 (a 22% gap).

Among the 97 participating named colleagues, civility ratings were 5.0/7 for those paired with a reinvestment employee, 5.2/7 for persistent-repair, and 4.2/7 for withdrawal (a 19% gap).

4. Discussion

An earlier CfAHC brief report found that volunteers who declined extra requests beyond their capacity were rated more reliable and were more often still active at six months. The pattern here is related, with one difference. Reinvestment employees did not save effort. They moved it, and the first two months of that move coincided with more tiredness, not less.

What they moved it toward appears to matter more than what they moved it away from. Withdrawal employees cut the same contact and gained little. A relationship that cannot be fixed is less like a leak to be plugged than a field left fallow: the gain comes from what gets planted elsewhere in the meantime.

The civility ratings bear on whether reinvestment is a polite word for shutting someone out. Colleagues on the other side of reinvestment rated their treatment close to those whose counterpart was still actively trying, and well above those whose counterpart had pulled back without reinvesting. Task-required contact continued in both reduced-contact groups, so the difference may lie in how that contact was conducted by someone who was no longer depleted by it. The design cannot confirm this.

The benefit may also not have stayed with the employee alone. Managers who did not know the groups rated reinvestment employees' collaboration with the wider team higher, and each new recurring contact was, by definition, someone else's working relationship too; whether those colleagues gained from it was not measured.

5. Limitations and Future Directions

Employees chose their own response, and those able to build new contacts quickly may have been more resilient to begin with. Exhaustion and engagement were self-reported by people who knew which path they had taken. The difficult relationship was identified by one side only; we did not assess whether the named colleague saw the relationship as difficult. The 97 participating named colleagues were a self-selected 23% of the 418 invited and may have been those on better terms with the employee; the subgroups behind the civility comparison are also small (28–36). Excluding the 45 partial or mixed responders may make the groups look more distinct than they are. Managers did not know the groups but could see who held new recurring meetings, and collaboration with the wider team partly overlaps the reinvestment definition, so that rating may partly restate how employees were classified. We collected no measures from the new contacts themselves. Six months may be too short to see whether reduced contact hardens into a permanent rift. A follow-up will offer a structured reinvestment plan to a randomly selected half of employees who report a difficult relationship, with twelve-month outcomes.

Editorial Note

This report underwent CfAHC's Rapid Review Protocol, including AI-assisted literature cross-checking and a two-reviewer human sign-off.