Top 50 Data Analyst Interview Questions 2026 — With Answers
Real SQL, Python, case study and behavioural questions asked at Google, Amazon, Flipkart, Swiggy, Paytm and 30+ top companies in 2026. Complete answers, difficulty ratings, and the companies that ask each question.
SQL Interview Questions — Data Analyst 2026
Definition: SQL (Structured Query Language) is used to query and manage data in relational databases. It is the #1 required skill for data analysts — tested in 95%+ of interviews.
Easy SQL Questions
| # | Question | Topic | Asked At |
|---|---|---|---|
| 1 | Write a query to find all customers from Mumbai with orders over ₹5,000 | WHERE + JOIN | Infosys, TCS |
| 2 | Find duplicate emails in a customer table | GROUP BY + HAVING | Wipro, Accenture |
| 3 | Get the top 5 products by revenue | ORDER BY + LIMIT | Meesho, CRED |
| 4 | Find all employees earning above the company average salary | Subquery | Amazon, Paytm |
| 5 | Count total orders and revenue per product category | GROUP BY + Aggregates | Flipkart, Swiggy |
WHERE filters individual rows before grouping. HAVING filters groups after GROUP BY runs. Use WHERE to filter raw data, HAVING to filter aggregated results. Example: WHERE salary > 50000 filters rows; HAVING COUNT(*) > 5 filters groups.
Medium SQL Questions
| # | Question | Topic | Asked At |
|---|---|---|---|
| 6 | For each department, find the employee with the highest salary | Window Functions: ROW_NUMBER() | Amazon, Razorpay |
| 7 | Find customers who ordered in Jan 2024 but NOT in Feb 2024 | NOT IN / Anti-JOIN | Flipkart, Swiggy |
| 8 | Calculate month-over-month revenue growth % | LAG() window function | Zomato, PhonePe |
| 9 | Find the percentage each category contributes to total revenue | SUM OVER() / PARTITION | Walmart, EY |
| 10 | Find each manager and the count of their direct reports | Self JOIN | Infosys, Accenture |
| 11 | Calculate cumulative revenue by date (running total) | SUM OVER ORDER BY | Meesho, CRED |
| 12 | Find the top 3 customers per city by revenue | ROW_NUMBER + PARTITION | Google, Amazon |
Hard SQL Questions
Full SQL practice: SQL Interview Hub with Live Compiler →
Python (Pandas) Interview Questions — 2026
Definition: Pandas is a Python library for data manipulation and analysis. It provides DataFrame objects for structured data operations. Pandas is tested in 75% of data analyst interviews at product companies.
| # | Question | Pandas Function | Difficulty |
|---|---|---|---|
| 13 | Find top 5 products by revenue using groupby | groupby + agg + sort_values | Easy |
| 14 | Merge two DataFrames and find average order value per city | merge() + groupby | Medium |
| 15 | Clean a DataFrame with missing values — explain and implement strategy | dropna, fillna, isnull | Medium |
| 16 | Write a function to detect outliers using IQR | NumPy + Pandas | Medium-Hard |
| 17 | Create a pivot table: revenue by region and product category | pivot_table() | Medium |
| 18 | Apply a custom function to transform a column | apply() + lambda | Easy |
| 19 | Calculate 7-day rolling average of daily active users | rolling() + mean() | Hard |
| 20 | Find users who churned (last purchase > 90 days ago) | pd.to_datetime + date arithmetic | Medium |
Business Case Study Questions
Case study rounds are asked at Google, Amazon, Flipkart, Swiggy, Zomato, and all major product companies. Three frameworks cover 90% of all case study questions:
Metric Drop Framework
Q: ‘DAU dropped 20% last Tuesday — how do you investigate?’ → Step 1: Clarify (what metrics, what time period, which platform?) → Step 2: Segment (by region, device, feature, user cohort) → Step 3: Hypothesise (external event? product change? data pipeline issue?) → Step 4: Validate → Step 5: Recommend fix
Metric Design Framework
Q: ‘How would you measure success of the new checkout feature?’ → Step 1: Define goal (increase conversion) → Step 2: Leading KPIs (checkout start rate, step completion rates) → Step 3: Lagging KPIs (GMV, revenue per user) → Step 4: Guardrail metrics (returns, cancellations) → Step 5: Dashboard structure
A/B Test Design Framework
Q: ‘Design an experiment to test if the new onboarding improves 30-day retention’ → Step 1: Hypothesis (new onboarding increases 30-day retention by 5%) → Step 2: Unit of randomisation (user_id) → Step 3: Sample size calculation → Step 4: Duration (≥1 full week, ideally 2 weeks) → Step 5: Success metrics + guardrail metrics → Step 6: Analysis plan
Behavioural Questions for Data Analyst Roles
- “Tell me about a time you used data to change a decision that was already made”
- “Describe a situation where your analysis was wrong — what did you do?”
- “How have you communicated complex data insights to a non-technical audience?”
- “Tell me about a project where you had to work with incomplete or messy data”
- “Describe a time you influenced a product decision with your analysis”
Answer all using the STAR method: STAR Method Complete Guide →
⭐ Key Takeaways
- SQL is tested in 95% of data analyst interviews — window functions are the hardest and most tested topic
- Python means Pandas: groupby, merge, pivot_table, apply, data cleaning are the core topics
- Three case study frameworks cover 90% of questions: metric drop, metric design, A/B testing
- Behavioural questions — prepare 5 achievement stories using STAR method
- Always clarify before solving a case study — structured thinking > speed
- Full free practice available: SQL Hub with live compiler and Python Hub with 20 pages
Practice these questions with a real mentor
Book a free data analyst mock interview. We run through SQL, Python and case study rounds with live feedback.
Book Free Mock Session