Short answer
Normalization organises tables to reduce redundancy and update anomalies. First normal form requires atomic values; second removes partial dependencies on part of a composite key; third removes transitive dependencies, so non-key columns depend only on the key. Denormalization deliberately reintroduces redundancy to speed up reads, usually in analytics or high-traffic read paths.
Example
Storing a customer’s address on every order row means an address change must update many rows. Moving it to a customers table and referencing it by id removes the duplication.
How to answer it in an interview
- Explain anomalies (insert, update, delete) as the motivation.
- Mention star schemas in data warehouses as intentional denormalization.