Summary of "Data Analyst🔥Technical Interview Questions ✅ | Top 15 Questions with Answers"

Data Analyst Technical Interview Questions (top 15) — Key points & preparation checklist

Speaker: Aarushi — data analyst at a mid‑size company who switched from software development and is sharing her personal interview experiences.
Video purpose: Shares the 15 most common technical (and a few related non‑technical) questions Aarushi was asked across multiple data‑analyst interviews, and how to prepare for them.
Scope: Focuses on technical topics (SQL, Excel, Statistics, Python). HR and scenario‑based questions are not covered in depth (Aarushi offers to make a separate video for those).

Prepare short answers for common non‑technical questions:
- Introduce yourself.
- Why do you want to become a data analyst?
- What are your strengths and weaknesses as a data analyst?
Practice hands‑on demos — some interviews give a laptop and ask you to demonstrate queries, joins, pivot tables, etc.
Memorize and practice core topics (SQL joins, pivot tables, VLOOKUP/HLOOKUP, etc.).
Prepare multiple methods/solutions for practical tasks so you can present options during interviews.

Many interviews focus on core, practical skills rather than advanced theory (but some companies will ask advanced Python questions).

WHERE vs HAVING
- WHERE filters rows before aggregation; HAVING filters groups after aggregation.
- Be ready to give examples and explain when to use each.
Types of JOINs and when to use them
- Know INNER, LEFT, RIGHT, FULL OUTER, CROSS JOIN (and possibly SELF JOIN).
- Practice writing join queries and explaining use cases.
Write a query to find the second highest sale by city
- Practice patterns: ORDER BY with LIMIT/OFFSET, window functions (ROW_NUMBER, RANK, DENSE_RANK), or subqueries to get the second max.
- Example: this type of question was asked of Aarushi at PolicyBazaar.
DELETE vs TRUNCATE vs DROP
- Understand differences in behavior, transaction/logging, side effects (constraints, identity resets, recoverability), and appropriate use cases.

Pivot Table
- Know how to build and explain pivot tables: which fields go into rows, columns, values, filters.
- Practice creating pivot tables on real datasets; interviews may ask for on‑the‑spot creation.
VLOOKUP vs HLOOKUP
- VLOOKUP = vertical lookup; HLOOKUP = horizontal lookup. Know example uses for each.
VLOOKUP/HLOOKUP: exact vs approximate match
- Explain the last parameter (TRUE/FALSE): TRUE = approximate (requires sorted key), FALSE = exact match.
Removing duplicates
- Know multiple methods: Remove Duplicates tool, conditional formatting + filter, formulas/helper columns, Power Query. Explain when to use each.
Conditional formatting
- Know how and why to apply it (highlight outliers, thresholds, duplicates, top/bottom values) and the basic steps.

Mean, Median, Mode
- Definitions, differences, and when to prefer each (e.g., use median for skewed distributions).
Standard deviation
- Definition and interpretation as a measure of dispersion; why it matters in analysis.
Correlation vs Causation
- Define both, explain the difference, give examples, and discuss the limitations of correlation (it does not imply causality).

Libraries commonly used for data analysis
- Be ready to name and describe: pandas, numpy, matplotlib, seaborn (and scikit‑learn if ML is relevant). Explain which functions/features you use most.
Common pandas functions
- Be prepared to discuss operations you use frequently: read_csv, head, info, describe, indexing/selecting, groupby, merge, pivot_table, apply, dropna/fillna, sort_values, agg, etc.
Pandas vs NumPy
- Know differences and use cases: NumPy for numerical arrays and performance; pandas for labeled tabular data and higher‑level manipulation.
Visualization in Python
- Explain when you use matplotlib vs seaborn; be able to justify library choice by the visualization need.
Plotting with pandas
- Know pandas.DataFrame.plot wrappers (they use matplotlib under the hood) for quick exploratory plots.
Handling missing data in Python
- Common methods: dropna, fillna (constant or forward/backward fill), interpolation, and advanced imputation (mean/median/model‑based). Be ready to explain tradeoffs and when to use each.
Python vs R
- High‑level differences: R is strong for statistical analysis and some visualizations; Python is widely used for general data science, ML, NLP, and integrates well with production systems. State your preference and rationale if asked.

Prepare and rehearse short answers for: intro, motivation, strengths/weaknesses.
SQL
- Practice WHERE vs HAVING examples.
- Memorize and practice all JOIN types with sample tables.
- Practice second‑highest (and top‑N) SQL patterns (window functions, subqueries).
- Review DELETE/TRUNCATE/DROP differences and use cases.
Excel
- Build pivot tables from sample data until fluent with rows/columns/values/filters.
- Practice VLOOKUP/HLOOKUP examples and exact/approximate match differences.
- Learn 2–3 ways to remove duplicates.
- Practice conditional formatting scenarios.
Statistics
- Review definitions and use‑cases for mean/median/mode and standard deviation.
- Prepare clear examples illustrating correlation vs causation.
Python
- List and be able to discuss libraries you use.
- Prepare examples of common pandas functions you use daily.
- Know when to use pandas vs numpy.
- Practice plotting with matplotlib/seaborn and pandas plotting.
- Be able to describe and implement basic missing‑data strategies.
Practice live demos (SQL queries, pivot tables, lookups) — interviews may require hands‑on tasks.

Aarushi — data analyst (video creator; main and sole speaker), sharing personal interview experiences and examples (including a question asked at PolicyBazaar).