Data analyst interview questions: what gets tested and how to answer
The analyst questions that decide the round — how you handle a metric drop, a stakeholder who wants a different answer, dirty data — plus what each is testing.
Analyst interviews are less about SQL syntax and more about judgement with incomplete data. The strongest signal is whether you can describe a method, a limitation and a decision in the same answer.
A key metric dropped 15% overnight. What do you do?
Tell me about a time your analysis changed a decision.
A stakeholder wants a different answer from your data. What happens?
How do you handle missing or dirty data?
How do you decide which metric to put in front of an executive?
Tell me about a time you were wrong about a number.
Questions people also ask
How much SQL is tested in an analyst interview?
Typically one technical screen with joins, window functions and a data-quality question. Beyond that, interviews test interpretation — the coding part is a gate, not the whole assessment.
Should I bring a portfolio?
Only if it demonstrates a decision, not a chart. One project described as problem, method, limitation and outcome is worth more than five dashboards.
Run a data analyst mock interview.
Questions one at a time, a five-category score, and a rewrite of every answer. Free during the beta.
Start practising