Business analysis glossary
Normalisation
What it is
Normalisation is organising a relational database to reduce duplicated data and avoid update errors, by splitting data into related tables so that each fact is stored once. The common levels are called first, second and third normal form, and most business systems aim for the third.
Why it matters
When the same customer address is stored in many tables, updates miss some copies and data becomes inconsistent. Understanding normalisation helps analysts read data models, understand why data is spread across tables and write correct joins. Reporting models are often deliberately less normalised for speed.
Example
An old order spreadsheet repeats the customer's name and address on every order row. When a customer moves, half the rows keep the old address. In the new design the analyst agrees a customer table and an order table linked by customer id, so the address is stored once.