Deduplication of Contact Data

What is Contact Data Deduplication ?

Contact data deduplication is the process of identifying and then eliminating duplicate contact records. Different matching levels can be used, dependent upon your needs. The most commonly used levels are:

How does it work ? 

I use software that utilises a combination of algorithms including fuzzy matching, error tolerance and phonetic matching. These take into account typos, spelling variations, misplaced information, abbreviations and nicknames. By allowing acceptable differences in your contact data, `non-exact’ as well as identical matches can be detected.

The deduplication programs I use are very flexible. They can be configured on multiple parameters (driven by the characteristics of your customer data and your requirements) so I can provide you with the very best results.

Even better results can be achieved by firstly cleansing addresses. Standardising & correcting customer postal addresses then makes it possible to find customer matches that you’d never thought existed, maximising the identification of duplicates.

If your data is of a sensitive nature, I can also present the matching results to you in a ranked table. Each duplicate group can then be manually verified as correct or otherwise before any customers are removed.

Because deduplication is such a key process for improving your data quality, it’s usually integral to Data Cleansing and Data Migration projects.

Enhanced Deduplication Techniques 

If you store additional customer details such as telephone numbers and email addresses, your deduplication results could be even further improved.

Using this information in the dedupe configuration could identify the same person at two completely different addresses. Such an approach can be especially useful if some of your contact data is of a `historical’ nature.

In above example, the two contact records have different postal addresses. However, the phone number and email address are the same and they fuzzy match on name. As the first entry is much more recent than the second, you could conclude that the latter entry relates to historical address and thus can be eliminated.

Benefits of Deduplication