Why joins are needed in SQL - Visual Breakdown

Concept Flow - Why joins are needed

Start with Table A

↓

Identify related data

↓

Combine matching rows

↓

Create joined result

↓

Use combined data

We start with two tables, find related rows, combine them, and get a new table with connected information.

Execution Sample

SQL

SELECT employees.name, departments.name
FROM employees
JOIN departments ON employees.dept_id = departments.id;

This query joins employees with their departments to show employee names alongside their department names.

Execution Table

Step	Action	employees row	departments row	Match Condition	Output row
1	Check employees row 1 with departments row 1	{id:1, name:'Alice', dept_id:10}	{id:10, name:'HR'}	10 = 10 (True)	{employee_name:'Alice', department_name:'HR'}
2	Check employees row 1 with departments row 2	{id:1, name:'Alice', dept_id:10}	{id:20, name:'Sales'}	10 = 20 (False)	No output
3	Check employees row 2 with departments row 1	{id:2, name:'Bob', dept_id:20}	{id:10, name:'HR'}	20 = 10 (False)	No output
4	Check employees row 2 with departments row 2	{id:2, name:'Bob', dept_id:20}	{id:20, name:'Sales'}	20 = 20 (True)	{employee_name:'Bob', department_name:'Sales'}
5	Check employees row 3 with departments row 1	{id:3, name:'Carol', dept_id:30}	{id:10, name:'HR'}	30 = 10 (False)	No output
6	Check employees row 3 with departments row 2	{id:3, name:'Carol', dept_id:30}	{id:20, name:'Sales'}	30 = 20 (False)	No output
7	No more rows to check	N/A	N/A	N/A	End

💡 All employees rows checked against all departments rows; join complete.

Variable Tracker

Variable	Start	After Step 1	After Step 4	Final
Current employees row	None	{id:1, name:'Alice', dept_id:10}	{id:2, name:'Bob', dept_id:20}	None
Current departments row	None	{id:10, name:'HR'}	{id:20, name:'Sales'}	None
Output rows	[]	[{employee_name:'Alice', department_name:'HR'}]	[{employee_name:'Alice', department_name:'HR'}, {employee_name:'Bob', department_name:'Sales'}]	[{employee_name:'Alice', department_name:'HR'}, {employee_name:'Bob', department_name:'Sales'}]

Key Moments - 2 Insights

Why do we need to check each row of one table against rows of another?

Because related data is stored separately, we must compare rows to find matches, as shown in execution_table steps 1-6.

What happens if no matching department is found for an employee?

That employee's row does not appear in the output, as seen in steps 5 and 6 where no matches produce no output.

Visual Quiz - 3 Questions

Look at the execution_table, what is the output row at step 4?

A {employee_name:'Bob', department_name:'Sales'}

B {employee_name:'Alice', department_name:'HR'}

C No output

D {employee_name:'Carol', department_name:'Sales'}

💡 Hint

Check the 'Output row' column in row 4 of execution_table.

At which step does the join process end?

A Step 6

B Step 5

C Step 7

D Step 4

💡 Hint

Look for the step with 'No more rows to check' in execution_table.

If employees had a dept_id that doesn't exist in departments, what happens to that employee in the output?

A They appear with empty strings

B They do not appear in the output

C They appear with NULL department name

D They appear multiple times

💡 Hint

Refer to steps 5 and 6 where no matching department means no output row.

Concept Snapshot

Joins combine rows from two tables based on a related column.
They match rows where the join condition is true.
Only matching rows appear in the output.
This lets us see connected data from separate tables.
Syntax example: SELECT * FROM A JOIN B ON A.key = B.key;

Full Transcript

Joins are needed because data is often split into different tables to keep it organized. To see related information together, we combine rows from these tables where their related columns match. For example, employees and departments are separate tables. By joining on department IDs, we get employee names with their department names. The process checks each employee row against each department row, outputs combined rows when IDs match, and skips rows without matches. This way, we get a new table showing connected data from both tables.