Wednesday, January 25, 2006

Cleanliness is next to Godliness ... Especially for Your Marketing Database

The goal of your database is to provide the information to help you eliminate excess marketing costs and to enhance revenue generation from improved customer targeting and retention. But for many marketers, the goal is elusive because of questionable data quality.

Your data is only as good as your data, or put another way that we are all familiar with – garbage in, garbage out. Data cleansing is the seemingly straightforward process of getting the garbage out and quality in. With so much importance riding on data quality, it is remarkable that this aspect of building a marketing database has so often been trivialized.

The database marketer needs to transform data from legacy systems into unique identifiable elements that are needed for building an accurate, integrated view of customers and the products they purchase or the services they use. Data cleansing involves such factors as: name-and-address standardization, gender assignment, geo-coding, ZIP+4 processing, postal coding (for foreign addresses), data correction and validation. Geo-coding is used to enhance addressability by verifying address integrity and assigning postal codes, or by assigning geographic codes to support geo-demographic mapping and site planning activities.

Before skipping this process consider the budget implications that bad data will wreak upon you into the future. If just 10% of your database addresses are incorrect and you have 500,000 prospective and current customers in the database, clearly 10,000 mailers will fail to reach your target market … sending out a $10 mailing to 10,000 bad addresses gives you a fast idea of the financial drain.

A clean, well integrated marketing database is mandatory for success in today’s competitive environment. You need to master data extraction, data transformation, data cleansing and eventually, data integration.

Gartner lists six major steps to cleaning data:

Elementizing: parsing data into components called “elements”

Standardizing: applying a standard form to these elements

Verifying: examining the elements and checking for errors

Matching: detecting identical elements such as the same address or name

Householding: Searching for information that indicates the same household

Documenting: capturing the results of the earlier steps in metadata to facilitate future data cleansing

In the end the task requires a combination of technology, software, processes and services to build a single, accurate and useful view of your prospective and current customers.


Post a Comment

<< Home

Site Feed
Enter your email address below to subscribe to Context Rules Marketing!

powered by Bloglet