How to answer “Tell me about a conflict” in an interview

The question that arrives with no warning

Ife is interviewing for a project coordinator role. Twenty minutes in, the panel asks: “Tell me about a time you disagreed with a colleague.” She had planned to say she is a good communicator who gets along with everyone. Now she has to give an actual example, and the only ones she can think of either make her look bad or make someone else look bad.

This is not an unusual trap. The question is built to be hard to answer with a slogan, and the first thing people reach for is a label.

Why interviewers ask for a past event

The US Office of Personnel Management describes behavioral description questions as asking applicants to describe a past behavior shown in a situation relevant to the competency of interest, and says they rest on the principle that past behavior is the best predictor of future behavior. That is official guidance on how federal interviews are designed, so a particular employer may differ.

The same page says interviews with more structure show higher validity and better agreement between raters. A review of the structured-interview literature by Levashina and colleagues summarizes twenty years of research and concludes that much is known but many questions remain open. Neither source tells you what a panel will think of your answer. They do explain why a concrete event is more useful to them than a trait.

There is a second reason to keep to events. In a large meta-analysis of feedback, Kluger and DeNisi found that interventions improved performance on average, but over a third decreased it, and that effectiveness fell as attention moved toward the self and away from the task. That study is about feedback, not interviews. Applying it here is our inference: describing the issue rather than someone’s character keeps the listener’s attention on what happened.

Choose the story before you choose the words

What follows is practical guidance, not a research-backed formula. A workable story usually has a disagreement about something checkable, such as a deadline, a division of work or a method, and an outcome you can describe honestly even if it was partial.

A meta-analysis of team conflict by De Dreu and Weingart separated conflict about relationships from conflict about tasks and found task conflict negatively related to team performance and satisfaction. A later, larger meta-analysis (De Wit, Greer and Jehn, 2012, 116 studies) did not find a strong negative link for task conflict, so the evidence on task conflict is mixed. For you, the useful part is the distinction: a story about a task disagreement is easier to tell without insulting anyone.

• Situation: where you were, and what the shared goal was, in a sentence or two.
• Events: what each of you did or said, as things a camera could have recorded.
• Effect: what it did to the work or the timeline.
• Your move: what you asked, offered or changed, including a question you put to the other person.
• Outcome and learning: what happened, and one thing you would do differently.

An answer made of labels

This answer is not dishonest, but it gives the panel nothing to check. “Difficult”, “lazy” and “resolved it” are conclusions, and the listener cannot tell what happened or what Ife did. It also hands the colleague a character verdict, which can leave the panel wondering how she would talk about them.

An answer made of events

The stronger answer is longer, and every sentence is something that happened. It names her own part, including that she assumed something she did not know. It describes a request, not a victory. You can swap in your own story, and you should: this one is invented for illustration, and a borrowed story will fall apart under follow-up questions.

If the follow-up gets harder

Panels often follow with “What would you do differently?” or “What if he had refused?” You do not need a perfect answer. You need to stay with what you know.

If you blank, you can say so and take a breath. If the question touches a real grievance, such as a discriminatory remark, a safety issue or a manager who ignored you, you do not owe the panel the details. You can say the matter went through the proper channel and describe your part.

If the panel says nothing after your answer, that is normal; many interviewers are writing. Wait, and ask if they would like more detail.

What the evidence does not show

The sources describe how structured interviews are designed, what a literature review concluded, and what meta-analyses found about feedback and team conflict. None shows that this answer structure improves your odds of an offer, and the feedback and conflict studies were not about interviews. Ife, Kwame and the venue project are invented for this example. Leetsoul has not been shown to cause better interview or career outcomes.

Some interviewers are unstructured or inconsistent, and a very good answer can still miss. If you are asked about protected characteristics or something unlawful in your region, you can decline to answer or ask how the question relates to the job. Rules vary by region, so check your local employment authority if you are unsure.

The linked practice gives you a different disagreement to describe in terms of events.