A unique identifier will be selected that will relate to both sets of data. The simplest way is to use the identifier CustomerID as it should guarantee that a unique number value will be given to each customer and will auto increment for new customer accounts. Once this is in place the next step is to look at all the combined accounts and check their indexes. Even though the data entered may be similar, not having the same index structures for all tables and columns will create problems when using update statements and other queries. Indexes are used to place constraints on the tables and columns, the most common are UNIQUE, NOT NULL, and FOREIGN KEY. When these constraints are applied to SQL, every time that a statement is created in the application it will check against these constraints before any changes are allowed to be made. This helps also prevent human errors from occurring when entering or removing data. The UNIQUE constraint ensures duplicate rows and table names will not be created. Even though the data entered may not be the same, having the same indexes for tables and columns will create problems when using update statements and other queries. The NOT NULL constraint is commonly combined with other constraints to specify when to allow or deny the use of NULL values. There are plenty of times when NULL values are used for non-applicable attributes. However, when a NULL value is not expected it …show more content…
Table data should not depend on anything other than a table 's primary key, data that does can be removed and placed in their own tables. Automobile, boat, RV, and home, should not exist in the same table as customer, these values rely on policy types as their own primary key to define their meanings. Policy types will then be moved to their own table and new primary key will be created and this becomes a foreign key in the customer table. The last step is removing dependencies, which is data that relies on more data to define itself. 3NF takes all the previous forms and removes all redundant and duplicate data. There is now a separate table for customer, policy type, claims, prior insurance company, etc. These all become foreign keys in the customer table that allows for reduction of a lot of wasted space in the database. The merged company data should now be fully cleaned and accurate enough to use for