When you create new databases, you will often hear phrases such as

1. This database really needs to be normalized
2. We need to split up these tables.
3. This table needs to be dissected into component parts.

These are referring to a methodology that makes using relational tables much easier and more flexible. At the same time, this approach is most efficient, eliminating duplications in the information you are storing in the system.
This fact is key to why a normalized database structure will be used, in most of the work you do with SQL.
When you normalize a table, you take out the duplicate information and place it in a related table. Consider the example in the next lesson.