What Is Database Normalization? Differences Between 1NF, 2NF, and 3NF

Database Normalization

Purpose

Eliminate data redundancy; reduce update, insert, and delete anomalies

First Normal Form (1NF)

• Each column must hold atomic values (no repeating groups or arrays)
• Example: Split phone1, phone2, phone3 into a separate phones table

Second Normal Form (2NF)

• Satisfies 1NF
• Non-key columns must depend on the entire composite primary key (no partial dependencies)
• Example: OrderItem(OrderID, ProductID, ProductName) — ProductName only depends on ProductID; extract it

Third Normal Form (3NF)

• Satisfies 2NF
• No transitive dependencies between non-key columns
• Example: Employee(ID, DeptID, DeptName) — DeptName depends on DeptID, not EmployeeID; extract it

Denormalization

Intentionally introducing redundancy for query performance (e.g., caching computed results)

• Normalization → less redundancy, better for writes
• Denormalization → more redundancy, better for reads (fewer JOINs)