Data Analyst Interview Questions
10 curated questions with evaluation guidance for hiring managers.
How would you approach analyzing a sudden 20% drop in user engagement? Walk me through your investigation process.
Look for a structured framework: segment the data (time, geography, device, user cohort), check for external factors (holidays, outages), form hypotheses, and validate with data. Avoid jumping to conclusions.
Explain the difference between correlation and causation with a real-world example from your work.
Should give a clear example and explain why establishing causation requires controlled experiments or advanced techniques. Red flag if they treat correlation as proof of causation.
You have a dataset with 30% missing values in a key column. How do you handle this?
Should discuss multiple strategies: deletion (listwise, pairwise), imputation (mean, median, regression, KNN), and when each is appropriate. Look for understanding of how missing data can bias results.
How do you design a dashboard that will be used by non-technical stakeholders?
Should emphasize understanding the audience, key metrics over vanity metrics, clear labeling, progressive disclosure, and actionable insights. Bonus for mentioning user testing with actual stakeholders.
Write a SQL query to find the top 5 products by revenue for each category. Explain your approach.
Should use window functions (ROW_NUMBER or RANK with PARTITION BY). Look for clean syntax, handling of ties, and awareness of performance implications on large datasets.
How do you ensure the accuracy and reliability of your analysis before presenting it to leadership?
Should mention data validation, cross-referencing with other sources, peer review, sanity checks against known benchmarks, and documenting assumptions. Thoroughness is key.
Explain A/B testing to a non-technical colleague. How do you determine the right sample size?
Should simplify without losing accuracy. Look for mention of statistical significance, confidence intervals, minimum detectable effect, and practical considerations like test duration.
How do you handle conflicting data from different sources?
Should discuss data lineage, source reliability, reconciliation processes, and documenting discrepancies. Look for a methodical approach rather than simply picking one source.
Describe a situation where your analysis changed a business decision. What was the impact?
Look for clear storytelling: problem, data approach, insight, recommendation, and measurable outcome. Strong candidates tie their work to business metrics (revenue, cost savings, retention).
What tools and techniques do you use for data cleaning? How do you handle outliers?
Should mention specific tools (Python/Pandas, SQL, Excel) and techniques (IQR method, Z-scores, domain knowledge). Look for judgment about when outliers are errors vs. genuine data points.
Want AI-generated interview questions tailored to your specific job description? Workro analyses your JD and generates behavioural and technical questions calibrated for the role, seniority level, and required skills — in seconds.
Try free