Interrogate the data before you argue with the conclusion. Check what the metric actually measures, over what time window, and whether it would capture the harm you are worried about. Then make your concern concrete and measurable: name the guardrail metric you expect to move. Propose a test or a staged rollout rather than a debate, and if you are overruled, disagree, commit, and instrument it.
Why interviewers ask this
Every designer eventually loses an argument to a number, and hiring managers want to see how you handle it. They are checking whether you can translate a user experience concern into business terms, whether you know that short-term metrics often miss long-term harm, and whether you can disagree productively and then support the decision instead of relitigating it for months.
How to structure your answer
- Ask what the metric measures and over what window before objecting.
- Translate your concern into a business risk with a named metric.
- Propose a guardrail that would detect the harm if it happens.
- Suggest a staged rollout instead of an argument about opinions.
- If overruled, commit, and make sure the risk is being watched.
Example answer
I start by understanding the number rather than fighting it, because usually the PM is not wrong, the metric is just narrow. So: what does this measure, over how long, and for which users. A lot of engagement metrics look great over two weeks and hide churn that shows up in month three. Then I make my worry concrete. Not 'this feels manipulative' but 'I think this increases signups by making cancellation harder to find, so I expect refund requests and one-star reviews mentioning billing to go up, and I think we will see it within a month'. Now it is a prediction rather than a preference, and we can both go and look. Then I propose the smallest way to find out, usually a staged rollout with that guardrail on the dashboard. If it goes ahead and I am still uneasy, I say my piece once, clearly, in writing, and then I help make it as good as it can be. What I will not do is quietly design it badly to prove a point, which I have seen and which is poison.
Walking into this interview soon? GhostPilot listens to your live call, spots the question the moment it is asked, and puts a structured answer on your screen in real time. Try it on your next mock, or grab a $29 Session Pass, no subscription, for the real thing.
See how it worksFollow-up questions to expect
- What would make you escalate rather than commit?
- How do you tell the difference between a user harm concern and personal taste?
- Tell me about a time you were overruled and it turned out fine.
Related ux designer questions
Your interviewer will ask their own version of this. Paste your actual job description into the free Question Predictor and get the 20 questions that role is most likely to ask, with what each one is really probing.
Predict my questions