How to Find the Nth Highest Salary in SQL: A Beginner’s Guide

May 12, 2025

If you’re just stepping into SQL or prepping for a job interview, you’ve probably come across this common question: How do you find the nth highest salary in a table? It may seem tricky at first, but with a clear understanding of SQL's ranking and filtering features, the solution becomes much more approachable.

In this beginner’s guide, we’ll explore various methods to retrieve the nth highest salary in SQL using simple examples and practical queries.

Why Is This Question So Popular?

This question helps interviewers assess your ability to:

Use SQL’s sorting and filtering features
Work with subqueries or window functions
Understand and apply ranking logic

The “nth highest” concept can be applied to many real-world scenarios beyond salaries—like finding the nth top-selling product or nth most-viewed article—making it a valuable skill.

Scenario Setup

Let’s say you have a table called employees like this:

id	name	salary
1	Alice	70000
2	Bob	85000
3	Charlie	60000
4	Dave	85000
5	Eva	90000

Now, let’s explore different ways to find the nth highest salary.

Method 1: Using `LIMIT` with `OFFSET` (MySQL, PostgreSQL)

A simple and intuitive approach if you're using MySQL or PostgreSQL.


SELECT DISTINCT salary
FROM employees
ORDER BY salary DESC
LIMIT 1 OFFSET N-1;

Example: 3rd Highest Salary


SELECT DISTINCT salary
FROM employees
ORDER BY salary DESC
LIMIT 1 OFFSET 2;

This query sorts the salaries in descending order, skips the first two, and then returns the next one — the 3rd highest.

Good for: Simple scenarios with databases that support LIMIT.

Method 2: Using `DENSE_RANK()` (SQL Server, PostgreSQL, Oracle)

This method is more robust, especially when handling duplicate salaries.

SELECT salary
FROM (
  SELECT salary, DENSE_RANK() OVER (ORDER BY salary DESC) AS rank
  FROM employees
) AS ranked
WHERE rank = N;

Example: 2nd Highest Salary

SELECT salary
FROM (
  SELECT salary, DENSE_RANK() OVER (ORDER BY salary DESC) AS rank
  FROM employees
) AS ranked
WHERE rank = 2;

This will return 85000 — even if it appears more than once, because DENSE_RANK groups equal values with the same rank.

Good for: When you want to return all entries sharing the same salary rank.

Method 3: Using a Correlated Subquery (Generic SQL)

This method works across many SQL engines, even if they don't support advanced functions.


SELECT DISTINCT salary
FROM employees e1
WHERE N - 1 = (
  SELECT COUNT(DISTINCT salary)
  FROM employees e2
  WHERE e2.salary > e1.salary
);

Example: 1st Highest Salary


SELECT DISTINCT salary
FROM employees e1
WHERE 0 = (
  SELECT COUNT(DISTINCT salary)
  FROM employees e2
  WHERE e2.salary > e1.salary
);

This counts how many salaries are greater than each one, and selects the row where that count is exactly N-1.

Good for: SQL engines without window functions.

Common Pitfalls to Avoid

Duplicate salaries: Be sure to use DISTINCT or DENSE_RANK() to handle them correctly.
NULL values: These can affect your sorting, so filter them out if needed.
Zero-based offset confusion: Remember, OFFSET 0 gives the highest, OFFSET 1 gives the second highest, and so on.

Conclusion

Finding the nth highest salary in SQL is a foundational skill that teaches you how to think critically with queries. By understanding and applying methods like LIMIT & OFFSET, DENSE_RANK(), or correlated subqueries, you'll be well-equipped to tackle variations of this problem in any SQL environment.

Search This Blog

AWS Insights