About Data Analyst interviews in India
Data analyst interviews in India test three things in roughly equal measure: hands-on SQL, the statistical judgement to draw correct conclusions, and the communication skill to turn a finding into a decision. Most rounds open with a live SQL exercise, move into a discussion of a real project on your CV, and close with a stakeholder scenario. Strong candidates lead with business impact rather than methodology, are honest about the limits of their data, and can switch between Excel, Python and a BI tool without losing the thread of the question.
🎯 Data Analyst Interview Tips
SQL DrillsPractise joins, window functions and aggregation live — most analyst rounds include a SQL test.
Tell the StoryLead with business impact and a clear chart, not the method or the raw numbers.
Stats BasicsBe solid on A/B testing, significance, sampling and handling missing data.
Tool ExamplesHave a real project ready for Excel, Python/Pandas and a BI tool.
🔧 Technical Questions
Technical Question 1
Write a SQL query to find the second highest salary in a table.
💡 How to answer: SELECT MAX(salary) FROM employees WHERE salary < (SELECT MAX(salary) FROM employees). Alternatively use DENSE_RANK() window function.
Technical Question 2
What is the difference between INNER JOIN and LEFT JOIN?
💡 How to answer: INNER JOIN returns only matching rows. LEFT JOIN returns all rows from left table, NULLs for non-matching right rows. Draw a Venn diagram mentally.
Technical Question 3
Explain the difference between GROUP BY and HAVING.
💡 How to answer: GROUP BY aggregates rows. HAVING filters those aggregated groups. WHERE filters before aggregation. Common interview trap.
Technical Question 4
What are window functions in SQL? Give an example.
💡 How to answer: ROW_NUMBER(), RANK(), DENSE_RANK(), LAG(), LEAD(), SUM() OVER(). Example: calculate running total of sales by region.
Technical Question 5
How do you handle missing data in a dataset?
💡 How to answer: Identify pattern: MCAR/MAR/MNAR. Options: drop rows, impute with mean/median/mode, forward fill for time series, flag as separate category.
Technical Question 6
What is the difference between correlation and causation?
💡 How to answer: Correlation: two variables move together. Causation: one causes the other. Ice cream and drowning — correlated but no causation. Use A/B tests to establish causation.
Technical Question 7
Explain cohort analysis and when you would use it.
💡 How to answer: Grouping users by shared characteristic at a point in time (e.g., signup month) and tracking behaviour over time. Used for retention, LTV, churn analysis.
Technical Question 8
What is A/B testing? How do you determine sample size?
💡 How to answer: Hypothesis test comparing control vs treatment. Sample size depends on: baseline conversion rate, minimum detectable effect, desired power (80%), significance (p<0.05).
Technical Question 9
What Python libraries do you use for data analysis?
💡 How to answer: Pandas (data manipulation), NumPy (numerical), Matplotlib/Seaborn (visualisation), Scikit-learn (basic ML). Describe a real project for each.
Technical Question 10
How would you identify outliers in a dataset?
💡 How to answer: Z-score (>3 std deviations), IQR method (1.5×IQR), box plots, scatter plots. Then decide: remove, cap, or investigate the outlier.
🧠 Behavioural Questions
Behavioural Question 1
Walk me through a data project you did end to end.
💡 How to answer: Business problem → data collection → cleaning → analysis → visualisation → insight → action taken → measurable outcome. Quantify impact.
Behavioural Question 2
Tell me about a time your analysis was wrong.
💡 How to answer: Own it, explain how you caught the error, what you changed in your validation process. Shows intellectual honesty and process improvement.
Behavioural Question 3
How do you explain complex data to non-technical stakeholders?
💡 How to answer: Lead with business impact, not methodology. Use simple charts. Tell a story: problem → insight → recommendation. Avoid jargon completely.
Behavioural Question 4
How do you prioritise multiple urgent data requests?
💡 How to answer: Impact × urgency framework. Communicate timelines upfront. Loop in manager for business priority decisions. Always acknowledge all requesters.
💡 Situational Questions
Situational Question 1
A dashboard you built is showing data that contradicts the CEO's assumption. What do you do?
💡 How to answer: Verify your analysis from multiple angles first. Then present the data transparently with context. Show multiple cuts. Be confident but diplomatic.
Situational Question 2
You receive raw data with no documentation. How do you start?
💡 How to answer: Explore shape, dtypes, null counts, unique values. Understand business context first by talking to the data owner. Document assumptions as you go.
Situational Question 3
You are asked to build a report in 2 hours that normally takes 2 days. What do you do?
💡 How to answer: Clarify minimum viable output. Build the 80% version fast. Communicate what is and isn't included. Deliver on time with caveats documented.
💰 Salary Questions
Salary Question 1
What is your expected CTC?
💡 How to answer: 'I'm targeting a package competitive with the market for this level of SQL, Python and dashboard experience. What is the budgeted range?'
Salary Question 2
We cannot match your current CTC.
💡 How to answer: Counter with: market data from Naukri/LinkedIn Salary Insights for your skillset, specific high-impact projects you have delivered, growth trajectory value.
🎤 Ask Interviewer Questions
Ask Interviewer Question 1
What does the current data stack look like — warehouse, BI tool, orchestration?
💡 How to answer: Shows technical curiosity and helps you assess if you'll be building vs maintaining.
Ask Interviewer Question 2
What are the top 3 business questions the analytics team is focused on this quarter?
💡 How to answer: Shows business alignment and tells you if data is strategic or just reporting.
Ask Interviewer Question 3
How does the team handle data quality issues?
💡 How to answer: Reveals data governance maturity, a key indicator of how frustrating or productive your day-to-day work will be.