If you find duplicates within your data, don't just delete them! Make a copy of any duplicate record by storing it separately in another database. If you perform these duplicate checks regularly, say once a month, this file will inevitably grow into a substantial suppression file (also known as a stopfile). This will enable you to cross reference it against any new prospect or customer data you add to your database in the future, therefore, preventing a common mistake - identifying duplicates only to effectively add it back in from another data source at a later date!
Where possible store your customer contact name details in their separate, constituent parts. ie. Title, forename, surname. This will make identifying duplicates easier, both for you or any professional company employed to do so. Also, as an added bonus it can cut down on reformatting costs as the name does not have to be split.
Try to keep all of your basic contact information, ie; address 1 -3, town, county, postcode, email, and telephone in the same uniform format. For example, try to avoid storing house or flat numbers in a separate field in one set of data but within address line 1 in another set.
Be sure of what net file you actually want! Do you want only one contact per household/company if there are multiple people, or all contacts so long as they are unique? If husband and wife are stored as two separate records do you want these merged into one record or contacted separately? Making the correct choice can give huge cost savings in wasted postage, brochures etc…
Always assign a URN (unique reference number). Not only will this enable you to export only the fields you need when the data needs cleansing, it will also mean new data can be tied back up with the original allowing you to accurately update your database