Customer Retention: Advanced Data-Matching Algorithms Improve Customer Recognition

Multichannel marketers’ customer files change rapidly with new information being contributed on a daily basis from stores, catalogs and the Web. Even with the most strenuous efforts in place to qualify new-to-file transactions, your customer file no doubt continues to accumulate disparate and seemingly un-related transactions.

These dynamic factors make it difficult to correctly identify and value each customer with traditional identity consolidation processes. Most merge/purge processes were developed more than 20 years ago and were never conceived to recognize the fluidity of movement, name change and channels in which customers interact today. Even the most advanced de-duplication processes use character-based logic and look-up tables that are ill-equipped to assess the totality of a customer’s name and address permutations that accumulate through multiple customer interaction channels. These processes easily are deceived by minor variations in the name and address elements such as married/maiden names, nicknames, typos and mis-keys. A typical file will contain 2 percent to 5 percent unidentified duplicate customers after a standard merge/purge process.

In recent years, however, advanced customer recognition solutions have emerged to address the dynamic nature of customers and the multiple methods of data collection. Such methods are powered by advanced data-matching algorithms and are being applied to customer recognition challenges for the purpose of identifying and collapsing the identities of customers, regardless of name and address permutations, omissions or mis-keys.

No single algorithm can efficiently and effectively power a matching technology due to the multiple culprits of customer identity and data quality error. Advanced solutions incorporate not one, but several advanced matching algorithms, each designed to group records into temporal data sets for the purpose of bringing visibility to distinct patterns of repetitious error. Once patterns of error are identified, the records can be referenced to consumer data sources, allowing for the remediation of the error and recognition of the true identity of the customer.

An Example

Following is a traditional processing view for four seemingly unrelated single-buying customers:
Customer Key #: 60608923
Channel: Store
Name: Beth Snyder
City/State: Denver, Colo.
Recency: 20040910
Dollars Spent: $180
Customer Key #: 60607244
Channel: Web

Name: Elizabeth Allen-Snyder
City/State: Denver, Colo.
Recency: 20040104
Dollars Spent: $210
Customer Key #: 60606135
Channel: Catalog

Name: Beth Allen-Snyder
City/State: Boulder, Colo.
Recency: 20030415
Dollars Spent: $260
Customer Key #: 60606004
Channel: Store

Name: Elizabeth Allen
City/State: Newark, N.J.
Recency: 20030108
Dollars Spent: $120

But an advanced processing view sees one consolidated customer who bought four times. For example:

ID Link: 00058
Customer Key: 60608923
Name: Beth Snyder
City/State: Denver, Colo.
Frequency: 4
$ Spend: $770

The result is a correct identification and valuation of the customer, no wasted mail costs, and a reprieve from the embarrassment of revealing to your customera through duplicate promotions that you really don’t know who they are.

Rod Ford can be reached at (866) 243-7883 or rford@cognitivedata.com.