🧠 WHY `GROUP BY` Exists

Let’s start with an example.

Suppose you have this table 👇

emp_id	emp_name	salary	department
1	Abhishek	90000	Engineering
2	Ravi	75000	Engineering
3	Soham	65000	Marketing
4	Amit	90000	Finance
5	Sneha	80000	HR
6	Kiran	70000	Engineering
7	Om	55000	Marketing

Now, if you want to find

“Total salary in each department”

You can’t just run:

SELECT department, SUM(salary) FROM employees;

⚠️ This fails — because the database doesn’t know how to combine multiple rows for each department.

So we tell the DB:

“Hey, group all rows having same department together, and then apply SUM on that group.”

That’s what GROUP BY does.

⚙️ HOW `GROUP BY` WORKS INTERNALLY

Think of it as a three-phase process inside the database engine: