30 Interview Questions for Data Analysts (Technical, SQL & Business Acumen)

Every company says they are "data-driven," but most are not — because they do not have the right people translating data into decisions. A great data analyst does not just write SQL queries. They understand the business, ask the right questions, find patterns others miss, and communicate findings in a way that drives action.

The biggest hiring mistake is over-indexing on technical skills and under-indexing on business acumen and communication. A candidate who can write perfect SQL but cannot explain what the data means to a non-technical stakeholder will not move the needle for your company.

These 30 questions are balanced across technical skills, analytical thinking, business understanding, and communication. Adjust the technical depth based on whether you are hiring a junior analyst or a senior lead. For instant custom questions, try our free AI interview question generator.

SQL is the foundation of data analysis. These questions test practical ability, not textbook knowledge.

1. Write a query to find the top 5 customers by total revenue in the last 12 months, excluding refunds.

What to look for: Correct use of GROUP BY, SUM, date filtering, WHERE clause for exclusions, ORDER BY with LIMIT. Bonus: asking about the schema before writing the query.

2. Explain the difference between INNER JOIN, LEFT JOIN, and FULL OUTER JOIN. When would you use each?

What to look for: Clear, practical explanation with real use cases — not just textbook definitions. Understanding of how NULLs appear in different join types.

3. How would you identify duplicate records in a table?

What to look for: GROUP BY with HAVING COUNT(*) > 1, understanding of what constitutes a "duplicate" (all columns vs. specific key columns), and approach to resolution.

4. What is a window function? Give an example of when you would use one.

What to look for: Practical examples — running totals, rankings, moving averages, row-over-row comparisons. Understanding of PARTITION BY and ORDER BY within window functions.

5. You have a query that takes 30 seconds to run. How do you optimize it?

What to look for: EXPLAIN/EXPLAIN ANALYZE approach, index awareness, avoiding SELECT *, reducing subqueries, and understanding of query execution plans.

6. What tools and languages do you use for data analysis beyond SQL?

What to look for: Python (pandas, numpy) or R for complex analysis, Excel/Sheets for quick work, and BI tools (Tableau, Looker, Power BI, Metabase) for visualization. Depth over breadth.

These questions reveal how candidates approach data problems and think about analysis.

7. Walk me through how you would investigate a sudden 30% drop in daily signups.

What to look for: Structured approach — verify the data first, check for technical issues (tracking, deployment), segment the drop (by source, device, geography, user type), compare to external factors, and form hypotheses before diving into data.

8. How do you distinguish correlation from causation in your analysis?

What to look for: Understanding of confounding variables, time precedence, controlled experiments (A/B tests), and humility in stating conclusions. A good analyst qualifies their findings.

9. Describe an analysis you did that changed a business decision. What was the impact?

What to look for: A concrete example with measurable impact. The story should show that they understood the business context, not just the data. Bonus: they followed up to measure the outcome.

10. How do you handle missing or messy data?

What to look for: Assessment of why data is missing (random vs. systematic), appropriate handling methods (imputation, exclusion, flagging), and transparency about data quality limitations in their findings.

11. You are asked to determine if a new feature increased user engagement. How do you approach this?

What to look for: A/B test analysis if available, before/after comparison with controls for seasonality, definition of "engagement" metrics, statistical significance awareness, and segment analysis.

12. How do you decide which metric to track for a given business question?

What to look for: Connecting business goals to measurable outcomes, understanding of leading vs. lagging indicators, awareness of gaming/manipulation risks, and simplicity preference.

The best analysts are business partners, not just query runners. These questions test that ability.

13. A product manager asks you: "How many users do we have?" How do you respond?

What to look for: Clarifying questions before answering — active users vs. registered, time period, specific product/feature, and the business context behind the question.

14. How do you present data findings to non-technical stakeholders?

What to look for: Lead with the insight (not the methodology), use clear visualizations, tell a story with the data, anticipate questions, and provide actionable recommendations — not just numbers.

15. Tell me about a time a stakeholder disagreed with your analysis. How did you handle it?

What to look for: Openness to being wrong, willingness to re-examine assumptions, clear methodology documentation, and professional discussion of differing interpretations.

16. How do you prioritize when multiple teams are asking for analysis simultaneously?

What to look for: Impact assessment, communication of timelines, ability to offer quick-and-dirty analysis vs. deep dive as appropriate, and escalation when needed.

17. What questions do you ask before starting a new analysis project?

What to look for: What decision will this inform? Who is the audience? What is the deadline? What data is available? What has already been tried? These questions prevent wasted work.

18. How do you ensure your dashboards and reports are actually used by the team?

What to look for: User research (what do stakeholders actually need?), simplicity, regular review and pruning of unused reports, and embedding data into existing workflows rather than creating separate destinations.

19. What makes a good dashboard? What makes a bad one?

What to look for: Good: answers a specific question, appropriate chart types, clear labels, minimal clutter, actionable. Bad: too many metrics, confusing visuals, no context, decorative rather than informative.

20. When would you use a bar chart versus a line chart versus a scatter plot?

What to look for: Bar for comparison between categories, line for trends over time, scatter for relationships between two variables. Understanding of when to use (and not use) pie charts.

21. How do you handle building a report that needs to serve both executives and operational teams?

What to look for: Layered approach — executive summary on top, drill-down details below. Different views or pages for different audiences rather than one cluttered report.

22. Walk me through how you built a dashboard or report you are proud of.

What to look for: Clear understanding of the business need, thoughtful metric selection, design choices explained, stakeholder feedback incorporated, and measurable impact on decisions.

23. Explain p-values in plain language. When do they matter and when do they not?

What to look for: Clear explanation without jargon, understanding of sample size effects, awareness that statistical significance does not equal practical significance, and appropriate skepticism about p-hacking.

24. How would you design an A/B test? What sample size would you need?

What to look for: Hypothesis formation, control vs. treatment groups, metric selection, sample size calculation (based on effect size and power), runtime estimation, and awareness of common pitfalls (peeking, multiple testing).

25. What is survivorship bias? Give a real-world example.

What to look for: Clear explanation with a relevant business example — analyzing only successful customers while ignoring churned ones, or looking at companies that survived while ignoring failures.

26. How do you handle outliers in your analysis?

What to look for: Investigation first (are they errors or genuine?), appropriate handling based on context (removal, capping, separate analysis), and transparency about the decision.

27. What is the most interesting pattern or insight you have discovered in data?

What to look for: Genuine curiosity and excitement about data. The story should show initiative (finding something unexpected) rather than just completing assigned tasks.

28. How do you stay current with new tools and techniques in data analysis?

What to look for: Practical learning habits — online courses, Kaggle, reading engineering blogs, experimenting with new tools, and community involvement.

29. What is your biggest data analysis mistake? What did you learn?

What to look for: Honesty and self-awareness. Common real mistakes: wrong date filter, misunderstood metric definition, survivorship bias, or presenting results without sufficient validation.

30. Where do you want to be in your data career in 3 years?

What to look for: Alignment with your growth path — do they want to go deeper technically (data engineering, data science) or broader (analytics management, product analytics)?

Structure Your Data Analyst Interview

We recommend this interview loop:

1. Phone Screen (30 min) — Analytical thinking and business acumen questions
2. Technical Assessment (60 min) — SQL problems and a take-home or live data analysis exercise
3. Case Study (45 min) — Present a business scenario and have them walk through their analytical approach
4. Culture & Communication (30 min) — Stakeholder communication, growth, and team fit

The case study is the most revealing round — it shows how candidates think about real problems, not just how they write SQL.

Need custom data analyst interview questions? Our free AI interview question generator creates tailored questions in seconds. Start building your full hiring pipeline with BorovaHR — free forever.