Home › Interview questions › Technology & IT › Data AnalystDATA ANALYST INTERVIEW PREP
Data Analyst Interview Questions (SQL, Excel, Business) with Tips
Data analyst interviews test SQL fluency, statistical sense and — most of all — whether you can turn data into a business decision.
What is the difference between WHERE and HAVING? Explain the types of SQL joins. Sales dropped 15% last month. How would you investigate? How do you explain a technical finding to a non-technical stakeholder? What is the difference between correlation and causation? How would you design an A/B test? Which dashboard are you most proud of? How do you handle missing or messy data?
1. What is the difference between WHERE and HAVING?
Why they ask: A quick SQL fundamentals check.
How to answer:
WHERE filters rows before aggregation; HAVING filters groups after GROUP BY. Give a one-line example, e.g. customers with more than five orders.
2. Explain the types of SQL joins.
Why they ask: Joins are the core of analyst SQL.
How to answer:
INNER, LEFT, RIGHT, FULL OUTER; mention anti-joins (LEFT JOIN … WHERE key IS NULL). Warn about row duplication from one-to-many joins and how you check for it.
3. Sales dropped 15% last month. How would you investigate?
Why they ask: The classic business case — tests structured thinking.
How to answer:
Confirm the data is right (tracking, definitions, late data). Segment: region, product, channel, new vs returning customers. Check external factors: seasonality, pricing changes, competitors, outages. Summarise the likely driver and a recommended action.
4. How do you explain a technical finding to a non-technical stakeholder?
Why they ask: Communication is half the job.
How to answer:
Lead with the answer and the decision it supports. One clear chart, plain language, no jargon. State confidence and caveats briefly.
5. What is the difference between correlation and causation?
Why they ask: Statistical judgment.
How to answer:
Correlation is co-movement; causation needs a mechanism and evidence. Confounders explain many spurious correlations. A/B tests or natural experiments are how you get closer to causation.
6. How would you design an A/B test?
Why they ask: Increasingly common for product analysts.
How to answer:
Hypothesis and primary metric. Sample size / power, randomisation unit, test duration. Guardrail metrics; avoid peeking; interpret significance and practical impact.
7. Which dashboard are you most proud of?
Why they ask: Shows product sense and impact.
How to answer:
Who used it and what decision it supported. Key design choices (few KPIs, filters, data freshness). The change it drove.
8. How do you handle missing or messy data?
Why they ask: Real-world data quality skills.
How to answer:
Find out why it is missing first. Options: remove, impute, flag — and the bias each introduces. Document assumptions so results are reproducible.
More interview guides Any job title works in CareerCue — these are just a starting point. See all fields →
Mind going blank in the real interview? CareerCue works for any job in any field. It listens to the question and shows a short cue based on your resume and the role, so you can answer in your own words. Free to try, no credit card.
Try CareerCue free →
CareerCue