The primary difference between the most complex variable data printing jobs and the simplest mail merge applications is the database. The more information housed within the database, the more complex a variable data application can be. Gathering additional information about an individual customer and storing it in the database allows a variable data document to be quite detailed in its personalization and elicits a much greater possibility for a customer response.

The variable data is housed within a database and is extracted from the database at the time the job is printed. The extraction is accomplished with the help of database publishing tools for the insertion of variable information onto almost any type of print application.

Some applications require so many variations in the information presented that complete changes in the layout, text, and graphics are required. For example, do-it-yourself manuals offered by a building supply company may have many different versions for the same type of project, such as building a deck. Changes in the instructions, supply lists, and illustrations are required for each style of deck while the basic format for each manual may be similar.

The data is stored in tables where each record has a number of fields containing information relating to that record. The information in the fields can be accessed for use in many areas, such as direct mail, variable imaging, and detailed reports. The data can be in the form of text or images.

Creativity is the driving force behind successful variable data applications used for direct marketing. Imaginative design and clever incorporation of the database within the application is what separates the best and most successful direct mail campaigns from the run-of-the-mill applications. Of course, the developer and/or printer of the direct marketing scheme can be quite limited with their alternatives in application development due to the meager database they are given to work with, but even a limited database can result in a unique and successful application with the help of a little ingenuity.

The table containing the data that is to be entered into the database should be broken into small units, which make up the fields for each record in the table. Breaking the information into small units will allow for greater flexibility when it is time to extract information from the database for printing and report analysis. Typical data units are items such as the customer's first name, customer's last name, address, state, zip, phone, and e-mail address. The data can then be queried by one or any number of the individual units.

The database will probably contain more than one table of information. A table organizes the information into rows and columns. The rows represent individual records and the columns are the individual fields for each record.

Using more than one table in a database makes it easier to organized all the information relating to a particular subject. A database developed for a company's orders may include tables such as a Customers table, Orders table, Products table, Payments table, and Shipping Information table.

Deciding which table should be used to store specific data is not always an easy task. If there is a one-to-one relationship between the fields, then they should go into the same table. For example, in the Customers table, each customer would have one address, one city, and one state, so the fields should be placed in the same table. The order and product information should not be placed in the Customers table because each customer may place more than one order and each order may contain more than one product. The orders should be in a separate table and since each order may include more than one product, the products should be in a table of their own.

Information can be pulled from several tables to be printed on the same form. The example below shows an invoice that would pull its information from three different tables. The customer's name and account number are pulled from the Customers table, the usage from the Detailed Usage table, and the invoice information from the Invoices table. The image would vary from invoice to invoice according to each record in the Invoice table.

The content of the fields in the database can be text or images. The images must be named so that they are easily identified with the records that they are associated with. 

With digital printing, the image in the database should be the correct size to fit into the area where it is to be printed. If the images are not the correct size, they will have to be resized at the time of the print run, resulting in an increase in production time.

The addresses in the database need to be kept up to date by checking for validity. Validation can be performed manually, but it can be very time consuming. CASS (Coding Accuracy Support System) certified software is available that will perform many different functions related to updating the addresses in a database.

Performing database maintenance in-house can be very time consuming and costly. It requires sophisticated computer equipment and software, and personnel that understand all of the requirements. Many print suppliers that use complex databases for digital print applications have database experts on staff. Professional service bureaus can be used to ensure that the database is processed by the necessary certified software and that it will meet all of the USPS address requirements.

The production department should be supplied with a complete printout of the variable data fields that will be used from the database. The production department will check the field names in the variable data production program to ensure that they match exactly with the field names in the database.