Behavioural interviews · Module 05 of 12 · Level 2: The core questions

Conflict and disagreement

"Tell me about a time you disagreed with a colleague." Interviewers aren't checking whether you were right. They're checking how you treat people you disagree with, how you settle it, and what you do once it's decided, especially when it doesn't go your way.

1. The idea

Two climbers roped together disagree about the route. A bad partnership argues on the ledge until one gives up in a sulk. A good one looks at the map and the weather together, picks a route, and then both climb it fully, including the one who preferred the other way. Conflict questions ask which kind of partner you are.

The wordings vary: "a time you disagreed with your manager", "a conflict with a teammate", "someone who was hard to work with", "a time you pushed back". The interviewer listens for five things:

The disagreement was about the work, not the person. Two reasonable views about a design, a priority or a deadline.
You understood their side. You can state their reasons fairly, even now. That's the clearest sign of maturity.
You used evidence, not volume. Data, a prototype, a small experiment, a customer's words.
It was resolved, and you can say how. Agreement, a compromise, or a decision by the right person. Escalation is fine if it's done openly and late, not first.
The relationship survived. You still work well with them, and once the decision was made, you committed to it, even when you lost.

You don't have to winSome of the strongest conflict answers end with "we went with their approach, and I put my full effort behind it." Amazon names this Have Backbone; Disagree and Commit: challenge respectfully, then commit wholly once it's decided.

2. How it works

A real disagreement, resolved the way strong answers describe it. Watch the dispute move from opinion to evidence, to a decision, to commitment.

3. Try it yourself

Check your conflict story

Answer six questions about a conflict story of yours, or pick an example. The verdict shows what an interviewer is likely to write, and what to add.

4. What interviewers look for

Conflict is scored on how you treated the other person and how the disagreement got resolved, more than on who was right.

Signal	Weak	Good	Strong
Respect	The other person is the villain	Neutral description of their view	You state their reasons fairly, and what was right about them
Method	Argued harder, or went straight to the manager	Talked it through one to one	Found shared goals, then used data or an experiment to decide
Resolution	Unresolved, or "they eventually agreed with me"	A compromise	A clear decision, by the right person, that both sides could commit to
Aftermath	Not mentioned	"We're fine now"	You committed fully, even if you lost, and the relationship got stronger

5. A model answer

"Tell me about a time you disagreed with a senior colleague." An invented answer, shown part by part. Read it for the shape, then build your own from a true story.

SituationOur principal engineer wanted to move our reporting service to a new event-streaming platform before the holiday peak. I was the engineer who ran the service day to day.

TaskI thought the timing was risky: the move would land three weeks before our busiest period, with no rollback plan. But she had more context than me, and I needed to understand why she wanted it now.

ActionI asked her for 30 minutes and started by asking what problem the move solved. It turned out the old queue was close to its storage limit, which I hadn't known. So we agreed on the goal (survive the peak) and disagreed only on the route. I proposed we measure: I replayed last year's peak traffic against both setups in staging. The new platform handled it; the old queue would have hit its limit on day two. I said so in our team channel the same day.

ResultWe migrated, with a rollback switch I built, and the peak went through without an incident. She later asked me to own the rollout plan for the next two services.

LearningI was right about the risk and wrong about the timing. Now I ask "what problem does this solve, and what happens if we wait?" before I push back.

Why it scoresThe colleague is treated fairly, and the candidate admits a gap in what they knew. The dispute is settled by an experiment, not by seniority or volume. The candidate publicly backs the outcome, even though it went against their first view.

6. Remember it

Memory hook: "Same goal, different route, test it"Agree on what you both want, disagree only about how, and let evidence decide.

How this hook works: what “Same goal, different route, test it” means

What the slogan means, phrase by phrase

“Same goal”: Start by finding what you both want: a reliable launch, a happy customer. Most conflicts shrink once that's said out loud.
“different route”: The disagreement is now about method, not about who is right or senior. That's a problem you can solve together.
“test it”: Settle it with data, a prototype or a small trial, and commit to the result, whoever it favours.

Why this picture fits: Two hikers who both want to reach the summit before dark can look at the map together. Two hikers arguing about who's the better navigator never move.

The story → the concept

The hikers	The disagreement
Both want the summit before dark	The shared goal
The north ridge vs. the valley path	The two approaches
Checking the map and the weather	Data, a prototype, an experiment
Both taking the chosen path	Disagree and commit

Watch it on an example

"We both want the peak to go without an incident." (same goal)
"You want to move now; I want to wait. Can we check which one is riskier?" (different route)
"I replayed last year's peak in staging. Moving now is safer; I'll build the rollback switch." (test it, then commit)

The trap it saves you from

Telling a conflict story where you're the hero and the other person is the obstacle.

It still works when…

There's no data to collect. Then the "test" is the decision-maker: agree who decides, give them both views fairly, and commit to the outcome.

7. Common mistakes

A villain storyIf the other person sounds stupid or malicious, the interviewer wonders how you'll describe them one day. Give their reasons fairly.

"I've never really had a conflict"Interviewers hear this as a lack of self-awareness or of stakes. Everyone disagrees at work; pick a real one.

Escalating firstGoing to the manager before talking to the person reads as avoiding conflict. Escalate openly, and only after trying.

You always winIf every conflict story ends with them admitting you were right, it sounds unlikely. Show one where you committed to their decision.

A personal clashStories about personalities are hard to resolve with evidence. Choose a disagreement about the work.

8. Practice exercise (25 minutes)

Pick two conflicts (5 min). One where you won, one where you didn't. Both should be about work decisions.
Write their side (5 min). For each, write the other person's view in their words, as fairly as you can.
Find the shared goal and the evidence (10 min). What did you both want? What settled it?
Tell the one you lost (5 min). Two minutes, recorded. It's usually the stronger story.

9. Self-check quiz

What should a conflict story show about the other person?

Stating their view fairly is the clearest sign of maturity.

Which resolution method do interviewers value most?

Evidence turns an argument about people into a question about facts.

You argued against a decision and lost. Is it still a good story?

Disagree and commit is exactly what many rubrics look for.

Which disagreement makes the strongest story?

Work disagreements can be resolved with evidence.

10. Go deeper

Resource	Time	Why
Have Backbone; Disagree and Commit (Amazon Leadership Principles)	2 min	A company's own wording of how to disagree, then commit.
Be direct and respect your colleagues (Meta values)	2 min	The same idea in Meta's published values.
Module 06: Failure and feedback	Next	The question most candidates dread.