Understanding DENSE_RANK in SQL

Introduction

When working with databases and analyzing data, ranking records is very important for organizing information based on certain conditions. One ranking function called `DENSE_RANK()` is useful because it assigns ranks to rows without leaving any empty spaces or gaps. This guide explains what `DENSE_RANK()` is, how it operates, and when to use it effectively in SQL.

Overview

• Understand the basic function and use of SQL’s DENSE_RANK() function.
• Use SQL queries using the DENSE_RANK() function to rank rows in a dataset according to predetermined standards.
• Manage scenarios where multiple records share the same ranking value and ensure consecutive ranking without gaps using DENSE_RANK().
• Implement ranking for statistical analysis, such as calculating percentiles and quartiles, ensuring a continuous sequence of ranks.
• Use DENSE_RANK() in conjunction with other SQL functions to produce detailed and insightful reports.

What is DENSE_RANK()?

The DENSE_RANK() function in SQL assigns a rank number to each row within a section or partition of the results. It works differently than the RANK() function, which may skip rank numbers when there are ties or identical values. With DENSE_RANK(), the ranks are assigned one after the other continuously, with no gaps. So if two rows have the same value and are tied for a rank, the very next rank number is used right after, without skipping any numbers.

SQL

DENSE_RANK() OVER (
    [PARTITION BY partition_expression]
    ORDER BY sort_expression [ASC | DESC]
)

• PARTITION BY: This optional clause divides the result set into partitions. The `DENSE_RANK()` function is applied to each partition separately. If omitted, the entire result set is treated as a single partition.
• ORDER BY: This clause specifies the order in which the rows are ranked.

How Does DENSE_RANK() Work?

To understand how `DENSE_RANK()` works, let’s consider an example. Suppose you have a table named `sales` with the following data:

| Product | Sales |

|---------|-------|

| A       | 100   |

| B       | 200   |

| C       | 200   |

| D       | 300   |

Using the `DENSE_RANK()` function to rank these products by their sales in descending order would look like this:

SQL

SELECT Product, Sales,
    DENSE_RANK() OVER (ORDER BY Sales DESC) AS Rank
FROM sales;

The result would be:

| Product | Sales | Rank |

|---------|-------|------|

| D       | 300   | 1    |

| B       | 200   | 2    |

| C       | 200   | 2    |

| A       | 100   | 3    |

As shown, products B and C have the same sales amount and are both ranked 2nd. The next rank is 3rd, without any gaps.

Practical Applications of DENSE_RANK()

`DENSE_RANK()` is particularly useful in various scenarios, such as:

• Identifying Top Performers: In business settings, you might need to identify top-performing salespeople, products, or departments. `DENSE_RANK()` can help you rank these entities without leaving gaps, providing a clear view of performance.

• Handling Ties: When multiple records share the same value, `DENSE_RANK()` ensures that they receive the same rank, and the next rank follows consecutively. This is useful in competitions or any scenario where tied results need to be handled gracefully.

• Pagination: In web applications, `DENSE_RANK()` can be used to implement pagination by ranking results and then displaying them in manageable chunks.

• Statistical Analysis: `DENSE_RANK()` is essential for various analytical functions, such as calculating percentiles, quartiles, and other statistical measures that require a continuous sequence of ranks.

Examples of DENSE_RANK() in Action

Let’s explore a few examples to illustrate the use of `DENSE_RANK()` in different contexts.

Example 1: Ranking Products by Price

Consider a `products` table with columns `product_id`, `product_name`, and `price`. To rank products by their price in descending order:

SQL

SELECT product_id, product_name, price,
    DENSE_RANK() OVER (ORDER BY price DESC) AS price_rank
FROM products;

This query will assign ranks to products based on their price, with the highest-priced product ranked first.

Example 2: Ranking Employees by Department and Salary

Suppose you have an `employees` table with columns `employee_id`, `department_id`, and `salary`. To rank employees within each department by their salary:

SQL

SELECT employee_id, department_id, salary,
    DENSE_RANK() OVER (PARTITION BY department_id ORDER BY salary DESC) AS salary_rank
FROM employees;

This query will rank employees within each department separately, ensuring that the ranking is based on their salary.

Differences Between RANK() and DENSE_RANK()

While both `RANK()` and `DENSE_RANK()` are used to rank rows based on specified criteria, they differ in handling ties:

• RANK(): Leaves gaps in the ranking sequence when there are ties. For example, if two rows tie for the first rank, the next rank will be 3.
• DENSE_RANK(): Does not leave gaps. The next rank will immediately follow the previous rank, even if there are ties.

Example:

Given the same `sales` table, using `RANK()` instead of `DENSE_RANK()`:

SQL

SELECT Product, Sales,
    RANK() OVER (ORDER BY Sales DESC) AS Rank
FROM sales;

The result would be:

| Product | Sales | Rank |

|---------|-------|------|

| D       | 300   | 1    |

| B       | 200   | 2    |

| C       | 200   | 2    |

| A       | 100   | 4    |

Notice the gap between ranks 2 and 4.

Conclusion

The `DENSE_RANK()` function is a useful tool in SQL for giving rank numbers to rows in a dataset based on certain conditions. The ranks will be one after the other, without any gaps, even if some rows have the same value and are tied. Understanding and using `DENSE_RANK()` can improve your ability to analyze data effectively and present it clearly. Whether you need to identify top performers, deal with ties or identical values, or do statistical analysis, `DENSE_RANK()` provides a solid way to rank data without leaving any empty spaces in the ranking sequence.

Frequently Asked Questions

Ayushi Trivedi

My name is Ayushi Trivedi. I am a B. Tech graduate. I have 3 years of experience working as an educator and content editor. I have worked with various python libraries, like numpy, pandas, seaborn, matplotlib, scikit, imblearn, linear regression and many more. I am also an author. My first book named #turning25 has been published and is available on amazon and flipkart. Here, I am technical content editor at Analytics Vidhya. I feel proud and happy to be AVian. I have a great team to work with. I love building the bridge between the technology and the learner.