Understanding the problem
Nearly every CRM company and internal corporate IT department has taken a stab at solving the problem of data normalization. Unfortunately, no one has done it right! Why? Think about it: when you buy a CRM it is usually empty. If you import dirty data from an old CRM, the new CRM will be dirty.
Normalization cannot be dictated at the vendor level
CRM systems are not designed to normalize data. Why? A good CRM must deliver flexibility to each client implementation; normalization cannot be dictated at the vendor level. Therefore, it is left to each individual customer and each individual user to enter and import data in the way that they see fit. For the CRM vendor it is a lose-lose scenario. If they dictate a data format, whichever format they choose, be it verbose or abbreviated, someone will not be happy.
Massive duplicates and miskeyed data does not become a problem until after you start to use a CRM. About a year after the CRM is implemented is typically when the buyer realizes there is a huge problem with the information. Without some systematic way to start with and keep information clean, duplicates will be introduced and someone has an opportunity to make more $$ on professional services.
CRM Shield is a product from Broadlook Technologies that ends the cycle of Dirty Data