For example, say they store City, State & Zip Code information for clients of the same board with other customer information. In this approach, keep repeating City, State & Zip Code information for all customers in the same area. In lieu of storing the same information over & over again, they could normalize the information & generate a related table called the city. The "City" table could then store City, State, ZIP code, along with identifiers that relate back to the customer table, & they can eliminate those five columns of the Customer table & add the new column of identity. Normalization is the method of designing a information model to efficiently store information in a database. The finish result is that redundant information is eliminated & only the information related to the attribute is stored within the table. Whether you have heard about normalization or not, your database most likely follows a quantity of the rules, unless all your information is stored in the giant box. Let's take a look at the first five normal forms & rules for determining different forms here. Normalization rules are divided in various ways. People often refer to the third normal form (3NF) when talking about database design. This is what most database designers trying to accomplish: In the conceptual stage, the information are segmented & normalized as far as possible, but for practical purposes, these segments are changed during the evolution of model information. Various normal forms may be introduced in different parts of the information model to handle the one-of-a-kind situations that can be found.