How to Repair a Working Relationship After a Conflict

Abstract

This article is about how to repair a working relationship after a conflict, and what research can and cannot say about it. Three pieces of research bear on it. A meta-analysis of 175 studies or samples (Fehr, Gelfand and Nag, 2010) looked at what goes with forgiveness, and found the strongest links with empathy, apology, the offender's intent and the person's anger. Two studies (Kim, Ferrin, Cooper and Dirks, 2004) found that an apology repaired trust better than denial after a competence failure, while denial did better after an integrity accusation when later evidence showed innocence. A laboratory study (Schweitzer, Hershey and Bradlow, 2006) found that trust damaged by untrustworthy behaviour could be rebuilt, but trust damaged by deception did not fully recover. None of these studied finance staff. The three steps at the end are practical readings, not tested methods, and this article is not medical advice.

The meeting where you did not look at each other

Two weeks ago you disagreed with a colleague in front of the team. It was not shouting, but it was sharp, and some of it was fair on both sides.

Since then you have been polite. You copy each other on everything, and you both know the polite version is not the real one. You wonder how to repair a working relationship that still has to function every Tuesday.

How to repair a working relationship: what is the evidence?

The research cited here does not give a recipe for repair. It covers a large meta-analysis of forgiveness, two experiments on apology and denial, and a laboratory study of rebuilding trust. These studies used a mix of samples and tasks, and none was limited to finance. The next three sections cover what each one found. None of them tested the three steps at the end.

What a meta-analysis of forgiveness found

Fehr, Gelfand and Nag (2010) combined 175 studies or samples, with 26,006 participants in total, and looked at 22 variables related to forgiveness. They asked which of them go with a person forgiving someone who has wronged them.

The strongest links, in a range of about .4 to .5 in size, were with state empathy (r = .51), the offender's intent (.49), apology (.42) and the person's anger (.41). More empathy and an apology went with more forgiveness. Variables about the situation explained more than variables about the person's character. The evidence is correlational. This is summarised from the published abstract.

Read the two weeks of polite silence through this. What happened between you and how it was handled went with forgiveness more than who either of you is. This is an inference. A correlation does not show that an apology causes forgiveness, and the studies were not about colleagues in particular.

What two studies of apology and denial found

Kim, Ferrin, Cooper and Dirks (2004) compared an apology with a denial after someone was accused of a trust violation. They separated violations about competence from violations about integrity.

After a competence violation, trust was repaired better by an apology than by a denial. After an integrity violation, trust was repaired better by a denial than by an apology. The apology worked when later evidence showed guilt, and the denial worked when later evidence showed innocence. This is summarised from the published abstract, which does not give the sample.

Read the argument you had through this. It may matter whether the other person feels you were incompetent, or feels you were dishonest. This is an inference. The study does not support denying something that did happen, and it was an experiment on accusations, not a model of a team row.

What a laboratory study of trust repair found

Schweitzer, Hershey and Bradlow (2006) studied what restores trust once it has been damaged. Trust that was harmed by untrustworthy behaviour alone could be effectively restored after people saw a consistent series of trustworthy actions.

Trust harmed by the same behaviour together with deception never fully recovered. This held even when the deceived people received a promise and an apology and then saw consistent trustworthy actions. A promise to change significantly sped up recovery, but earlier deception reduced how much it helped. This is summarised from the published abstract.

Read the weeks ahead through this. Repeated, consistent small actions did more for trust than a single gesture. This is an inference. The study was a laboratory task, and the abstract says nothing about how long repair takes in a real workplace.

How to repair a working relationship: three steps

None of these is a tested method, and none was studied in finance teams. They combine what the research points to with ordinary practice. Where a step goes beyond the research, it says so.

1. Decide whether the issue is competence or honesty before you speak. The apology study found that the two call for different responses. If the issue is a competence failure, a plain apology fits what was found. If it is about honesty and the accusation is false, say plainly that it is false. This is a judgement, and no study cited here tested this step. For instance: "I got the numbers wrong on Tuesday and I should have flagged it earlier. I am sorry."

2. Ask for their view and listen before you explain yours. Empathy was among the strongest links with forgiveness, though the evidence is correlational. Asking is a way to find out what the other person saw. This is a judgement, and no study cited here tested this step. For instance: "I would like to hear how that meeting looked from your side before I say anything."

3. Offer one small, specific commitment and keep it every time. The trust study found that a promise sped up recovery, and that consistent trustworthy actions did the rebuilding. Pick something you can repeat without effort. This is a judgement, and no study cited here tested this step. For instance: "You will have my part of the pack by Wednesday noon, every week this month."

If the conflict is between two people you manage, How to Handle Conflict Between Two Team Members goes into that. For the conversation itself, there is Difficult Conversations: A Framework for Honest Communication. If the row started with a mistake, see How to Tell Your Boss You Made a Mistake in the Numbers. When what you need is for the other person to feel heard, The Art of Emotional Validation: Why "I Understand" Isn't Enough is about that. And How to Build Trust in a Team: What the Research Supports looks at the longer game.

Questions People Ask

Should I apologise even if I think I was partly right?

The apology study found that an apology repaired trust after a competence failure. You can apologise for your part without accepting the whole account. What to say is a judgement for you, and the studies cited here did not test it.

What if they will not talk to me?

The studies cited here do not cover that case. A short, specific invitation with no demand attached leaves the choice with them. How long to wait before trying again is a judgement, not a research finding.

How long does trust take to come back?

The laboratory study found recovery after a consistent series of trustworthy actions, but it does not give a time. Deception damaged trust more lastingly than untrustworthy behaviour alone. Real workplaces vary, and no study cited here measured a timeline.

Can coaching help me repair a relationship?

Coaching is educational support for work patterns. It can help you plan what to say and notice how you react when you meet. It is not treatment, and it cannot change how the other person responds. No study cited here tested coaching.

When it is more than a tense relationship

A tense relationship is one thing. It is more when the conflict involves bullying, harassment, threats or discrimination, or when you feel unsafe at work. In those cases talk to HR, your manager, your union or an employment adviser. If you also feel persistently low or anxious, cannot sleep, or notice physical symptoms, talk to your doctor or a qualified mental health professional. This article is not medical advice. Coaching is educational support for work patterns. It is not treatment and not a substitute for it.

Sources

• Fehr, R., Gelfand, M. J., & Nag, M. (2010). The road to forgiveness: A meta-analytic synthesis of its situational and dispositional correlates. Psychological Bulletin, 136(5), 894–914. DOI: 10.1037/a0019993 (Summarised from the published abstract.)

• Kim, P. H., Ferrin, D. L., Cooper, C. D., & Dirks, K. T. (2004). Removing the shadow of suspicion: The effects of apology versus denial for repairing competence- versus integrity-based trust violations. Journal of Applied Psychology, 89(1), 104–118. DOI: 10.1037/0021-9010.89.1.104 (Summarised from the published abstract.)

• Schweitzer, M. E., Hershey, J. C., & Bradlow, E. T. (2006). Promises and lies: Restoring violated trust. Organizational Behavior and Human Decision Processes, 101(1), 1–19. DOI: 10.1016/j.obhdp.2006.05.005 (Summarised from the published abstract.)