“Tell me about a conflict with a teammate.” “Tell me about a time you disagreed with your manager.” “Tell me about a time you pushed back.” These are among the most common behavioral questions, and among the easiest to fail, because the natural way to tell a conflict story is as a story about being right.
The interviewer does not care much who was right. They want to know what you are like to disagree with. This post covers what is actually being scored, the shape of a mature answer, the phrasings that quietly sink an otherwise good story, and a skeleton to build your own.
One note on the examples. The worked answers in this track come from real, solo projects, and solo projects have no teammates to clash with. So rather than invent a conflict, this post gives you the structure and contrasting phrasings, plus one real story of a different kind of disagreement: with a rule, using evidence. Build your teammate story from your own work, using the skeleton at the end.
What is being scored
THEY ARE LISTENING FOR RED FLAGS
you sought to understand the other side the other person is a villain
you argued from evidence, not authority "I was right and they eventually saw it"
you separated the person from the problem it got personal, or you went around them
you reached a decision, even if not yours it was never resolved; you just stopped
you committed once it was decided you were right, so you did it your way
the relationship survived "we didn't really talk after that"Two items on the left matter most. Evidence over authority: a strong answer moves the disagreement from opinions to something testable — data, an experiment, a prototype, a past incident. Commitment after the decision: if the decision goes against you, you back it fully. This is usually called disagree and commit, and saying you did it is worth more than winning the argument.
The shape of a mature answer
A good conflict answer moves through five beats:
- The disagreement, stated neutrally. What the two positions were, described so that the other side sounds reasonable. If your description of their view makes them sound foolish, the interviewer hears that you never understood it.
- Why it mattered. The stakes: a deadline, a risk, a cost. A conflict over something trivial is a story about stubbornness.
- How you engaged. Directly, privately first, curious before persuasive. The key move: what would change my mind, and what would change theirs?
- How it was decided. By evidence, an experiment, or the person who owns the decision. Name who decided.
- What happened after. The outcome, how you supported the decision, and the state of the relationship.
Notice that “I was proved right” is not one of the beats. If you were, it can appear in the outcome, briefly and without triumph. Some of the strongest conflict answers end with the candidate's view losing and the candidate explaining how they made the other approach succeed.
Phrasings that sink a good story
The same events can score very differently depending on the words. A few contrasts:
INSTEAD OF SAY
"he just didn't get it" "we were optimising for different things:
he for shipping this week, me for the
on-call load next quarter"
"I escalated to our manager" "we agreed we couldn't settle it ourselves,
so we took both options to the person
who owned the decision, together"
"eventually they came around" "the load test settled it"
"I went ahead and did it my way" "it was decided against me; I said so once,
then I made their approach work"
"she was being difficult" (describe what she wanted, and why it
was reasonable from where she sat)Escalation is worth a specific word. Going over someone's head without telling them is one of the fastest ways to lose points in this answer. Taking a disagreement together to the person who owns the decision is normal and healthy, and saying it that way shows you know the difference.
Disagreeing with your manager
The manager version adds a power imbalance, and the interviewer is now also checking two things: that you will push back when it matters (they do not want someone who silently executes a bad plan), and that you do it without undermining them.
The answers that score well share a pattern: the candidate raised the concern early, privately, with specifics — ideally a cost or a risk the manager had not seen — proposed an alternative rather than only objecting, and then accepted the manager's call. Phrases like “I wanted to make sure you had the full picture before deciding” capture the posture exactly. It is your job to make the information visible; it is their job to decide.
If the manager was ultimately right, say so. If they were wrong and you were right, say what happened in neutral terms and what you both took from it. Never tell this story in a way that makes your former manager look foolish. The interviewer is very likely to be your next one.
A real story: disagreeing with a rule, using evidence
Not every disagreement is with a person. Pushing back on a process, a standard or a requirement is a legitimate answer to “tell me about a time you pushed back”, and it exercises the same muscles: understanding why the rule exists, and arguing from evidence. This one is real.
Situation. I was rewriting a series of tutorials on my site. Some of the existing posts were very thin — a few hundred words — so I'd set a rule that every rewrite had to at least double the length of the page it replaced. There was a second rule, set for readers: every post had to fit in a 15 to 20 minute read, which caps it at about 4,400 words.
Action. The automated check failed one post. The longest existing page was already 2,871 words, and doubling it meant 5,742 — which the reading-time cap forbade. The two rules contradicted each other on exactly that one post. The easy move was to bend the post, or quietly disable the check. Instead I asked which rule was serving its purpose. The doubling rule existed to stop thin pages being replaced by thin pages; for that one post, the page wasn't thin. So the rule was wrong, not the post. I changed it: a rewrite has to reach the series' minimum length, which for the thin pages is still a doubling or more, and it can never come out shorter than the page it replaces.
Result. The post shipped inside the reading-time budget, and the rule still catches what it was meant to catch on every other page. I wrote down why the original rule was wrong, so nobody, including me, puts it back.
Learning. When a rule blocks good work, the first question is what the rule is for. Sometimes the work is wrong; sometimes the rule is.
What this story shows is the core of every good conflict answer, without any interpersonal drama: understanding the purpose behind the other position (here, the rule), arguing from a concrete number, changing the thing that was actually wrong, and recording the decision. When you tell a teammate story, those are the beats to hit.
A skeleton for your own
THE TWO POSITIONS mine: ____________ theirs: ____________
(write theirs so it sounds reasonable)
WHAT EACH OPTIMISED mine: ____________ theirs: ____________
STAKES what it would cost to get this wrong
HOW I ENGAGED privately / directly / curious first
"what would change my mind?" "what would change theirs?"
HOW IT WAS SETTLED evidence / experiment / the decision owner (name them)
OUTCOME what was decided, and whose view it matched
AFTER how I supported it; the state of the relationship
LEARNING one sentenceIf you are short of material, look for moments in code review, in design discussions, or in planning, where you and a colleague wanted different things. They do not need to be dramatic. A disagreement about a migration strategy, settled by a test run, is a better story than a shouting match, because it shows exactly the behaviour the interviewer is looking for.
What to take from this
- The question measures what you are like to disagree with, not who was right.
- Describe the other side so it sounds reasonable; move the argument to evidence.
- Name who decided, and show you committed to the decision either way.
- With a manager: raise it early and privately, bring an alternative, and accept their call.
- Never invent a conflict. If your bank is thin here, build the story from real team work before the interview.