Databases typically use number identifiers to ensure that each entry has at least one unique attribute. These identification numbers let users locate every entry in the database even if other fields don't contain unique information.
Organizations use databases to store nearly any type of information, and they allow users to search for information. Locating a particular record, however, can be difficult if different entries have the same information; for example, two people can share the same name, making it impossible to locate a specific record by name alone. By giving every entry a unique identification number, users can be assured that they can access the record they're looking for.
In many cases, users set the database to generate a unique number automatically. This number can be sequential, or users can set it to generate a random number. By letting the database choose a number, users can ensure that each number the database uses is truly unique.
Users can also enter their own unique identification number. Organizations might choose to use social security numbers, for example. However, doing so it potentially risky. Mistyping a number can cause the database to have two entries under what should be a unique identifier, and storing a social security number in a database can compromise a client's identity if someone gains unauthorized access to the database.