Method of analyzing credit card transaction data

• US 20080086365A1
• Filed: 10/05/2007
• Published: 04/10/2008
• Est. Priority Date: 10/05/2006
• Status: Active Grant

• Current Assignee: Pamela Tournier, Richard Zollino
• Original Assignee: Pamela Tournier, Richard Zollino
• Inventors: Tournier, Pamela, Zollino, Richard
• Granted Patent: US 8,812,351 B2
• Time in Patent Office: Days
• US Class Current: 705/10
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

A method of analyzing debit and credit card transaction data to provide interpretations of customer purchasing patterns for use by third parties, such as financial services marketers, in providing offers and incentives to targeted groups of consumers. Bulk credit card or debit card transaction data that has been gathered for a large sample of cardholders is obtained, a multi-step process is applied to prepare the data for analysis, multiple categories of marketing “intelligence” or decisions are attached to each transaction contained in the bulk transaction data, a summation of all of the different intelligence categories for each cardholder and for the entire sample is performed to create comparative normalizations, and a score in vector form is generated for each customer based on differences or variations in the way in which that customer shops relative to the bulk-derived normalized data for each field or dimension within the comparative normalization.

Claims

1. A method of analyzing customer purchase transaction data to provide targeted offerings to selected customers identified by the analysis as more likely than other customers to respond to the targeted offerings, wherein the transaction data comprises bulk transaction data relating to a multiplicity of purchase transactions effected by a plurality of customers with a plurality of merchants, the bulk transaction data comprising a multiplicity of records each relating to a purchase transaction by one of the plurality of customers and identifying the specific customer and the merchant of the purchase transaction, said method comprising the steps of:

associating each of said plural merchants with characterizing fields in a plurality of category taxonomies, said fields identifying possible characteristics of the plural merchants, by identifying which selected ones of said fields are relevant characterizers of each of said plural merchants;

correlating the plural bulk transaction data records with the associated characterizing fields of said plural merchants to generate, for each said taxonomy, a first weighted table of the purchase transactions at the plural merchants for each of said characterizing fields of the said each taxonomy;

correlating, for each of said plural customers, the transaction data records of said each of the plural customers with the associated characterizing fields of said plural merchants to generate, for each said taxonomy, a second weighted table of the purchase transactions of the said each of the plural customers at the plural merchants for each of said characterizing fields of the said each taxonomy;

comparing for said each of the plural customers, for each of said characterizing fields of the said each taxonomy, the weights of the first and second tables, to calculate for each of said characterizing fields of the said each taxonomy a difference of said weights of the first and second tables; and

identifying for said each of the plural customers, in said each taxonomy, from among the said calculated differences of said weights of the first and second tables for said plural characterizing fields, a greatest of said calculated differences;

said greatest of said calculated differences in each of said taxonomies identifying a characteristic of said each of the plural customers for use in targeted offerings to said each of the plural customers based on said identified characteristic.

2. A method in accordance with claim 1, wherein said taxonomies comprise one or more of “

motivations”

, “

lifestyles”

, “

location”

, “

media and marketing modalities”

, and “

ethnic and cultural tags”

.

3. A method in accordance with claim 1, wherein said first weighted table comprises, for said each taxonomy, a listing of said fields identifying possible characteristics of the plural merchants and for each said field a value identifying a weighted distribution for said each field representing, by said field characterizations, the multiple purchase transactions of said plural customers at said plural merchants.

4. A method in accordance with claim 1, wherein said second weighted table for said each customer comprises, for said each taxonomy, a listing of said fields identifying possible characteristics of the plural merchants and for each said field a value identifying a weighted distribution for said each field representing, by said field characterizations, the multiple purchase transactions of said each customer at said plural merchants.

5. A method in accordance with claim 3, wherein said second weighted table for said each customer comprises, for said each taxonomy, a listing of said fields identifying possible characteristics of the plural merchants and for each said field a value identifying a weighted distribution for said each field representing, by said field characterizations, the multiple purchase transactions of said each customer at said plural merchants.

6. A method in accordance with claim 5, wherein said comparing step comprises calculating for each of said characterizing fields of the said each taxonomy a difference of said weights by calculating, for each of said characterizing fields of the said each taxonomy, a difference between said identifying value in said first table and said identifying value in said second table.

7. A method in accordance with claim 6, further comprising the step of normalizing the differences calculated for the characterizing fields of the said each taxonomy to create a normalized distribution of said differences to further characterize purchasing characteristics of the said each customer.