What SQL questions are asked in TCS, Wipro, Infosys fresher interviews in 2026?

TCS, Wipro, Infosys, Cognizant, and Accenture fresher SQL interviews in 2026 consistently cover six topic clusters: (1) Basic SQL - SELECT, WHERE, ORDER BY, DISTINCT, LIMIT/TOP, aliasing; (2) Joins - INNER, LEFT, RIGHT, FULL OUTER, CROSS, SELF, with at least one written query; (3) Aggregates and GROUP BY - COUNT, SUM, AVG, MIN, MAX, HAVING, GROUP BY vs WHERE; (4) Subqueries - single-row, multi-row, correlated, EXISTS, IN, ANY, ALL; (5) DDL/DML - CREATE TABLE, ALTER TABLE, INSERT, UPDATE, DELETE, TRUNCATE, DROP, constraints (PRIMARY KEY, FOREIGN KEY, UNIQUE, CHECK, NOT NULL); (6) Real scenarios - find the second highest salary, find duplicates, find employees without managers, running totals. TCS NQT typically has 8-12 SQL MCQs plus 1-2 written queries. Wipro NLTH and Infosys InfyTQ include live SQL problems. Cognizant GenC adds harder subqueries and window functions. Capgemini and HCL focus on joins and basic aggregates.

What is the difference between INNER JOIN, LEFT JOIN, RIGHT JOIN, and FULL OUTER JOIN?

INNER JOIN returns only rows where the join condition matches in BOTH tables - rows without a match in either table are excluded. LEFT JOIN (or LEFT OUTER JOIN) returns ALL rows from the left table, with the right table's columns NULL where no match exists. RIGHT JOIN is the mirror image: ALL rows from the right table, NULLs from the left where no match exists. FULL OUTER JOIN returns ALL rows from both tables, with NULLs on the side that didn't match. Practical example: customers LEFT JOIN orders gives every customer (including those with no orders); INNER JOIN gives only customers who have placed at least one order. Most databases support all four; SQLite only supports LEFT JOIN explicitly (RIGHT and FULL are simulated by reversing the table order or UNION). Most fresher interviews ask you to write a query using all four against a small employees-departments schema.

How do you find the second highest salary from an employees table?

Several approaches, in increasing order of generality: (1) LIMIT/OFFSET: SELECT DISTINCT salary FROM employees ORDER BY salary DESC LIMIT 1 OFFSET 1 - simple and widely used; works in MySQL/PostgreSQL/SQLite. (2) Subquery: SELECT MAX(salary) FROM employees WHERE salary < (SELECT MAX(salary) FROM employees). (3) Window function (most modern): SELECT salary FROM (SELECT salary, DENSE_RANK() OVER (ORDER BY salary DESC) AS rnk FROM employees) t WHERE rnk = 2. (4) For ties handled: use DENSE_RANK instead of ROW_NUMBER so all employees with the second-highest salary are returned. Cognizant GenC and Capgemini frequently ask this question with the twist "handle ties" or "what about the top N with ties."

What is the difference between UNION and UNION ALL?

UNION combines the results of two SELECT statements and removes duplicate rows (effectively applying DISTINCT). UNION ALL combines the results and keeps all rows, including duplicates. UNION is slower because the database has to sort and deduplicate the combined result set; UNION ALL is faster. Use UNION when duplicates are not desired (e.g., combining lists of unique employee IDs from two sources); use UNION ALL when you know the result sets are already distinct or when duplicates don't matter (e.g., combining two log tables for analysis). Both require the SELECT statements to have the same number of columns with compatible data types. MySQL supports both; SQL Server supports both; older SQLite versions had limited UNION support but current versions are full.

What are SQL window functions and when would you use them?

Window functions perform a calculation across a set of rows that are related to the current row, without collapsing the result like GROUP BY does. They are written using the OVER() clause with optional PARTITION BY (groups to compute over), ORDER BY (ordering within each partition), and a frame clause ROWS/RANGE BETWEEN (which rows to include). Common window functions: ROW_NUMBER() (unique sequential number within partition, no ties), RANK() (rank with gaps for ties), DENSE_RANK() (rank without gaps for ties), NTILE(n) (divide into n buckets), aggregate functions used as window functions (SUM, AVG, COUNT, MIN, MAX with OVER), LEAD(column, n) and LAG(column, n) (access next/previous row's value), FIRST_VALUE and LAST_VALUE (first/last value in frame). Common interview scenarios: running totals (SUM over ORDER BY date), top N per group (ROW_NUMBER with PARTITION BY), year-over-year comparisons (LAG to get previous year's value), ranking within department (RANK with PARTITION BY department). Window functions are required for Cognizant GenC Next and Infosys Power Programmer roles; they are common at product companies (Flipkart, PhonePe) for analytics interviews.

What is the difference between DELETE, TRUNCATE, and DROP?

DELETE removes rows from a table one at a time (or in batches), can use a WHERE clause to filter which rows to remove, fires triggers, can be rolled back in a transaction, is slower than TRUNCATE for large tables, and uses the transaction log. TRUNCATE removes all rows from a table in one operation, cannot use WHERE (removes everything), does not fire triggers, cannot be rolled back in most databases (the operation is auto-committed and minimally logged), resets identity counters in some databases (SQL Server, MySQL with InnoDB). DROP removes the entire table definition and all its data from the database; the table no longer exists after DROP. Common interview question: which is faster, DELETE or TRUNCATE? TRUNCATE is faster because it logs minimally and deallocates entire data pages; DELETE logs each row deletion. Which to use depends on whether you need to filter (DELETE), need to keep the table but remove data (TRUNCATE), or need to remove the table entirely (DROP).

What is the difference between a primary key and a unique key?

Both PRIMARY KEY and UNIQUE KEY enforce uniqueness of values in a column or set of columns, but they differ in semantics and behavior. A table can have only ONE primary key but MULTIPLE unique keys. The primary key cannot have NULL values (in most databases, though SQL Server allows it but it's bad practice); unique keys allow one NULL value per column (in most databases). The primary key is typically the main identifier used for foreign key references; unique keys are used for additional uniqueness constraints (like email or phone number on a users table). The primary key creates a clustered index by default in SQL Server and MySQL/InnoDB (the table data is physically stored in the order of the primary key); unique keys create non-clustered indexes. Both automatically create indexes. Primary keys are typically used in joins and as foreign keys in related tables; unique keys are used to enforce business rules.

SQL Interview Questions 2026: Top 40 Freshers & Experienced

Q: What is the difference between WHERE and HAVING?

WHERE filters individual rows BEFORE aggregation, and cannot reference aggregate functions. HAVING filters groups AFTER aggregation (groups created by GROUP BY) and CAN reference aggregate functions like COUNT, SUM, AVG. Example: SELECT department, COUNT(*) FROM employees WHERE hire_date > '2020-01-01' GROUP BY department HAVING COUNT(*) > 5 - the WHERE filters out employees hired before 2020, then groups the remaining rows by department, then HAVING keeps only departments with more than 5 employees. WHERE is faster than HAVING because it reduces the number of rows before grouping. A query can have both WHERE and HAVING in the same SELECT: WHERE first, then GROUP BY, then HAVING. Common fresher trap: putting aggregate conditions in WHERE (like WHERE COUNT(*) > 5) - this is a syntax error; use HAVING instead.

Q: What is the difference between UNION and UNION ALL?

UNION combines the results of two SELECT statements and removes duplicate rows (effectively applying DISTINCT). UNION ALL combines the results and keeps all rows, including duplicates. UNION is slower because the database has to sort and deduplicate the combined result set; UNION ALL is faster. Use UNION when duplicates are not desired (e.g., combining lists of unique employee IDs from two sources); use UNION ALL when you know the result sets are already distinct or when duplicates don't matter (e.g., combining two log tables for analysis). Both require the SELECT statements to have the same number of columns with compatible data types. MySQL supports both; SQL Server supports both; older SQLite versions had limited UNION support but current versions are full.

Q: What are SQL window functions and when would you use them?

Window functions perform a calculation across a set of rows that are related to the current row, without collapsing the result like GROUP BY does. They are written using the OVER() clause with optional PARTITION BY (groups to compute over), ORDER BY (ordering within each partition), and a frame clause ROWS/RANGE BETWEEN (which rows to include). Common window functions: ROW_NUMBER() (unique sequential number within partition, no ties), RANK() (rank with gaps for ties), DENSE_RANK() (rank without gaps for ties), NTILE(n) (divide into n buckets), aggregate functions used as window functions (SUM, AVG, COUNT, MIN, MAX with OVER), LEAD(column, n) and LAG(column, n) (access next/previous row's value), FIRST_VALUE and LAST_VALUE (first/last value in frame). Common interview scenarios: running totals (SUM over ORDER BY date), top N per group (ROW_NUMBER with PARTITION BY), year-over-year comparisons (LAG to get previous year's value), ranking within department (RANK with PARTITION BY department). Window functions are required for Cognizant GenC Next and Infosys Power Programmer roles; they are common at product companies (Flipkart, PhonePe) for analytics interviews.

Q: What is the difference between DELETE, TRUNCATE, and DROP?

DELETE removes rows from a table one at a time (or in batches), can use a WHERE clause to filter which rows to remove, fires triggers, can be rolled back in a transaction, is slower than TRUNCATE for large tables, and uses the transaction log. TRUNCATE removes all rows from a table in one operation, cannot use WHERE (removes everything), does not fire triggers, cannot be rolled back in most databases (the operation is auto-committed and minimally logged), resets identity counters in some databases (SQL Server, MySQL with InnoDB). DROP removes the entire table definition and all its data from the database; the table no longer exists after DROP. Common interview question: which is faster, DELETE or TRUNCATE? TRUNCATE is faster because it logs minimally and deallocates entire data pages; DELETE logs each row deletion. Which to use depends on whether you need to filter (DELETE), need to keep the table but remove data (TRUNCATE), or need to remove the table entirely (DROP).

Q: What is the difference between a primary key and a unique key?

Both PRIMARY KEY and UNIQUE KEY enforce uniqueness of values in a column or set of columns, but they differ in semantics and behavior. A table can have only ONE primary key but MULTIPLE unique keys. The primary key cannot have NULL values (in most databases, though SQL Server allows it but it's bad practice); unique keys allow one NULL value per column (in most databases). The primary key is typically the main identifier used for foreign key references; unique keys are used for additional uniqueness constraints (like email or phone number on a users table). The primary key creates a clustered index by default in SQL Server and MySQL/InnoDB (the table data is physically stored in the order of the primary key); unique keys create non-clustered indexes. Both automatically create indexes. Primary keys are typically used in joins and as foreign keys in related tables; unique keys are used to enforce business rules.

Quick Answer: SQL Interview Questions 2026

The top 40 SQL interview questions asked at TCS, Wipro, Infosys, Cognizant, Accenture, Capgemini, and HCL in 2026 cover six clusters: basic SQL (SELECT, WHERE, ORDER BY), joins (INNER, LEFT, RIGHT, FULL OUTER, SELF), aggregates and GROUP BY (COUNT, SUM, AVG, HAVING), subqueries (single-row, multi-row, correlated, EXISTS), DDL/DML (CREATE, INSERT, UPDATE, DELETE, constraints), and real scenarios (second highest salary, duplicates, employees without managers, running totals). Most fresher interviews include 8-12 SQL MCQs (TCS NQT, Wipro NLTH) plus 2-3 written queries (Infosys InfyTQ, Cognizant GenC). Experienced interviews (3+ years) add window functions (ROW_NUMBER, RANK, LAG, LEAD, SUM OVER), CTEs, and query optimization (EXPLAIN, indexes). For the full Q&A collection with 25 curated questions, see the TutorsBot SQL Interview Questions hub.

The 6 Topic Clusters Asked Most Often

Cluster 1: Basic SQL (SELECT, WHERE, ORDER BY, DISTINCT)

Round 1 in nearly every interview. The expected baseline questions: What is the difference between DISTINCT and GROUP BY? How do you sort results (ORDER BY ASC/DESC, multiple columns)? How do you limit the number of rows returned (LIMIT n in MySQL/PostgreSQL/SQLite, TOP n in SQL Server, FETCH FIRST n ROWS ONLY in standard SQL)? How do you filter rows (WHERE with comparison operators, IN, BETWEEN, LIKE, IS NULL, AND/OR/NOT)? What is the difference between WHERE and HAVING? How do you alias columns and tables (SELECT col AS alias, FROM table alias)? TCS NQT and Wipro NLTH include 3-5 questions on this cluster; they are easy wins if you prepare.

Cluster 2: Joins (INNER, LEFT, RIGHT, FULL OUTER, SELF, CROSS)

Where most SQL interview time goes. The non-negotiable question: explain the difference between INNER, LEFT, RIGHT, and FULL OUTER JOIN with examples. Other classics: What is a self join (a table joined to itself, useful for employees-managers, employees-in-the-same-department)? What is a CROSS JOIN (Cartesian product, every row of A paired with every row of B)? What is the difference between JOIN and APPLY (SQL Server)? What is a natural join (joins on columns with the same name, rarely used in practice)? Most fresher interviews ask you to write 2-3 join queries against an employees/departments schema; Cognizant GenC Next adds harder multi-table joins (3+ tables with aggregations).

Cluster 3: Aggregates and GROUP BY

Tested heavily at Capgemini and Accenture. Questions: What aggregate functions are available (COUNT, SUM, AVG, MIN, MAX, plus database-specific ones like STRING_AGG, ARRAY_AGG, GROUP_CONCAT)? What is the difference between WHERE and HAVING (WHERE filters rows before grouping; HAVING filters groups after aggregation)? Can you use column aliases in GROUP BY (no in most databases - use the underlying expression)? What does COUNT(*) vs COUNT(column) return (COUNT(*) counts all rows including NULLs; COUNT(column) counts only non-NULL values)? What is the difference between GROUP BY and DISTINCT (they often produce the same result for a single column, but GROUP BY is required when you also want aggregates)?

Cluster 4: Subqueries

The classic differentiator between fresher and mid-level answers. Single-row subqueries return one value (used with =, <, >, etc.). Multi-row subqueries return multiple rows (used with IN, ANY, ALL). Correlated subqueries reference the outer query and run once per row (slower than non-correlated). EXISTS tests for the existence of rows in the subquery (returns TRUE/FALSE). Common interview questions: write a query to find employees whose salary is greater than the average salary in their department (correlated subquery or window function). What is the difference between a correlated and non-correlated subquery (execution frequency: correlated runs per outer row, non-correlated runs once)? When would you use a subquery vs a JOIN (modern SQL prefers JOIN for most cases; subqueries are still needed for EXISTS, NOT EXISTS, and complex aggregations)?

Cluster 5: DDL/DML and Constraints

Asked at TCS Digital and Infosys Power Programmer. Questions on CREATE TABLE (column definitions, data types, constraints), ALTER TABLE (ADD/DROP/MODIFY column, ADD/DROP constraint), constraints (PRIMARY KEY, FOREIGN KEY with referential actions ON DELETE CASCADE/SET NULL/RESTRICT, UNIQUE, CHECK, NOT NULL, DEFAULT), data types (INT, BIGINT, DECIMAL, FLOAT, VARCHAR, CHAR, TEXT, DATE, TIMESTAMP, BOOLEAN, JSON, BLOB), indexes (CREATE INDEX, when to use, B-tree vs hash, covering index), and views (CREATE VIEW, updatable views, materialized views). Common fresher question: explain the difference between CHAR and VARCHAR (CHAR is fixed-length, padded with spaces, slightly faster for known-length data; VARCHAR is variable-length, saves space).

Cluster 6: Real Scenarios (Second Highest Salary, Duplicates, Running Totals)

The most-tested questions across all companies. Five classics that come up in 80%+ of fresher SQL interviews: (1) Find the second highest salary - several approaches (LIMIT/OFFSET, subquery, window function with DENSE_RANK). (2) Find duplicate rows in a table - GROUP BY column HAVING COUNT(*) > 1, or window function ROW_NUMBER() OVER (PARTITION BY col). (3) Find employees who don't have a manager - LEFT JOIN employees e2 ON e.manager_id = e2.id WHERE e2.id IS NULL. (4) Delete duplicate rows keeping one - use ROW_NUMBER in a subquery to identify duplicates, then DELETE the rest. (5) Calculate running total - SUM(col) OVER (ORDER BY date). For experienced interviews, expect follow-ups like "what about ties in second highest salary" (use DENSE_RANK instead of LIMIT) or "what about when salary has NULLs" (filter NULLs in the WHERE clause).

Common Fresher Pitfalls to Avoid

Confusing WHERE and HAVING: WHERE filters rows, HAVING filters groups; aggregate functions are only allowed in HAVING (or in SELECT with GROUP BY)
Forgetting to alias aggregates: SELECT COUNT(*) gives a column named COUNT(*); use AS alias for readability
Confusing INNER and LEFT JOIN: INNER excludes unmatched rows; LEFT keeps all rows from the left table
NULL comparisons: = NULL never matches; use IS NULL or IS NOT NULL; COUNT(column) excludes NULLs, COUNT(*) includes them
Missing the GROUP BY column: Every non-aggregated column in SELECT must appear in GROUP BY
Confusing UNION and JOIN: UNION combines rows vertically (same columns); JOIN combines columns horizontally (different columns from different tables)
Not handling ties: LIMIT/OFFSET skips ties; use DENSE_RANK or RANK when ties matter
Subquery vs JOIN performance: Modern query optimizers treat subqueries and JOINs similarly; don't assume subquery is slower without EXPLAIN evidence

How to Prepare in 30 Days

The strongest 30-day fresher SQL interview preparation plan:

Week 1 - Foundations: Complete SQLBolt (free, 18 lessons), W3Schools SQL tutorial, write 20+ queries against a sample employees database (try SQL Murder Mystery or the Northwind sample database)
Week 2 - Joins and aggregates: Practice 30+ join queries, write the second highest salary in 4 different ways (LIMIT, subquery, window function, CTE), practice GROUP BY with HAVING and multiple aggregates
Week 3 - Subqueries and DDL: Practice correlated vs non-correlated subqueries, learn the difference between UNION and UNION ALL, write CREATE TABLE statements with all constraint types
Week 4 - Window functions and mock tests: Learn ROW_NUMBER, RANK, DENSE_RANK, LAG, LEAD, SUM OVER; take 2-3 timed mock tests on HackerRank SQL; review the SQL Interview Questions hub for the full Q&A collection

For a structured, project-based path from SQL basics to interview-ready, the TutorsBot Full Stack Development training covers database design, schema management, and API development end-to-end.

Frequently Asked Questions

How many SQL questions are asked in TCS NQT 2026?

TCS NQT 2026 typically includes 8-12 SQL questions in the coding/technical section for Digital and Prime profiles, plus 1-2 advanced SQL problems. For Ninja profiles, the count is lower (5-8 MCQs, 1 written query). TCS Digital adds harder scenarios including window functions and CTEs.

Is SQL enough to get placed in TCS/Wipro/Infosys?

SQL alone is rarely enough - companies test a combination of SQL, Python or Java, DSA, and aptitude/reasoning. The strongest fresher profile combines: SQL (intermediate to advanced, including window functions), one strong programming language (Python or Java), DSA (50-100 problems on LeetCode Easy/Medium), communication skills, and 2-3 small projects on GitHub. Cognizant GenC and Capgemini have the heaviest SQL testing of the major service companies.

What is the salary for SQL freshers at these companies in 2026?

Salary bands for freshers with SQL skills in 2026: TCS Ninja ₹3.36 LPA, TCS Digital ₹7-8 LPA, TCS Prime ₹9-11 LPA. Wipro ₹3.5-5 LPA. Infosys (InfyTQ Power Programmer) ₹6.5-9.5 LPA, regular InfyTQ ₹3.6 LPA. Cognizant GenC ₹4.5-6 LPA, GenC Next ₹7-12 LPA. Accenture AASE ₹4.5-6 LPA, AASS ₹6-9 LPA. Capgemini ₹4-6 LPA. HCL ₹3.5-5.5 LPA. Product companies (Flipkart, PhonePe, Razorpay) pay 2-3x these bands for SQL-heavy data analyst and analytics engineer roles.

What is the best resource for SQL interview preparation?

The strongest resources are: SQLBolt (free interactive lessons), W3Schools SQL tutorial (free reference), Mode Analytics SQL Tutorial (free advanced topics), HackerRank SQL domain (free hands-on practice), LeetCode Database problems (free Easy/Medium), the Northwind or Sakila sample databases for practice queries, and the TutorsBot SQL Interview Questions hub for curated 25 questions at all difficulty levels. For experienced interview prep, focus on window functions (the official PostgreSQL or SQL Server window function tutorial), query optimization (use EXPLAIN plans), and CTEs (Common Table Expressions).

Is SQL enough for data analyst interviews at product companies?

For entry-level data analyst roles at product companies (Flipkart, PhonePe, Razorpay, Cred), SQL is the core requirement - expect 3-4 SQL rounds covering joins, window functions, aggregations, and at least one complex business question. Strong SQL plus Python (pandas) and a data visualization tool (Tableau, Power BI, Looker) is the strongest entry-level profile. For senior data analyst or analytics engineer roles, expect SQL plus Python/R for statistical analysis, A/B testing knowledge, and a take-home assignment.

Resources and Next Steps

The authoritative sources listed (W3Schools, SQLBolt, Mode Analytics, PostgreSQL documentation, HackerRank) are the canonical references for SQL fundamentals and interview preparation. For the full 25-question curated Q&A collection at all difficulty levels, see the TutorsBot SQL Interview Questions hub. For related interview prep, see our Python interview questions, DBMS interview questions, and Normalization in DBMS guides.

#sql interview questions #sql interview questions and answers #sql interview questions for freshers #sql interview questions for experienced #sql interview 2026 #sql coding interview #sql joins interview

← Back to Blog

SQL Interview Questions and Answers 2026: Top 40 for