Customer identification system and method

Abstract

A method and system are presented for identifying a customer in a commercial transaction using less than complete identifying information. A name for the customer is extracted from a credit card during a purchase transaction. A trade area for the point of sale location is used to restrict a search of a demographic database to find a list of potential identity matches having names similar to the name on the credit card. A best match generator creates a profile of the expected purchaser of the products in the transaction. Using demographic information about each identity in the list of potential identity matches, each identity is compared to the profile and given a score. The highest scoring identity is then considered the best match. The best match identity is then assumed to be the identity involved in the transaction, and the customer database is updated to reflect this determination

1 . A computer implemented method for identifying a customer in a customer database system that tracks purchase transactions, the customer having purchased a product at a point of sale location by using a payment mechanism to pay for the product purchase, the method comprising: 
 a) determining a customer name for the customer from the payment mechanism;    b) creating a computerized list of potential matches, the potential matches each being an identity of a particular person, wherein an identity constitutes information that uniquely identifies a particular person, the list of potential matches being created by 
 i) searching in a computerized database for identities having a name similar to the customer name, with similar names being determined through at least one mechanism chosen from the set including identically matching names, nicknames, abbreviations, and known variations of the customer name, and  
 ii) combining a plurality of the identities found by searching the computerized database into the computerized list of potential matches; and  
   c) selecting a best match identity from the computerized list of potential matches, the best match identity indicating which of the plurality of identities in the list of potential matches is to be selected as the identity for the customer name; and    d) using the best match identity as the identity for the customer name in the customer database system.    
     
     
         2 . The method of  claim 1 , wherein the step of selecting a best match further comprises creating a score for each identity in the list of potential matches, and then selecting the identity with the highest score as the best match identity.  
     
     
         3 . The method of  claim 2 , wherein the step of creating a score for each identity in the list of potential matches further comprising running a plurality of tests each having a separate score, and combining the score for all tests to create a score for the identity.  
     
     
         4 . The method of  claim 3 , wherein one of the tests relates to whether the identity has previously purchased an item in the same category of goods as the product purchased at the point of sale location.  
     
     
         5 . The method of  claim 3 , wherein one of the tests relates to whether the name of the identity in the list of potential matches is a perfect match for the customer name determined from the payment mechanism.  
     
     
         6 . The method of  claim 3 , wherein one of the tests relates to the income of the identity in the list of potential matches.  
     
     
         7 . The method of  claim 3 , wherein one of the tests relates to the education level of the identity in the list of potential matches.  
     
     
         8 . The method of  claim 3 , wherein one of the tests relates to the age of the identity in the list of potential matches.  
     
     
         9 . The method of  claim 3 , wherein one of the tests relates to the age of children in the home of the identity in the list of potential matches.  
     
     
         10 . The method of  claim 3 , wherein the plurality of tests includes: 
 i) a first test that relates to whether the identity has previously purchased an item in the same category of goods as the product purchased at the point of sale location,    ii) a second test that relates to the income of the identity in the list of potential matches, and    iii) a third test that relates to the age of the identity in the list of potential matches.    
     
     
         11 . The method of  claim 10 , further comprising a fourth test that relates to whether the name of the identity in the list of potential matches is a perfect match for the customer name determined from the payment mechanism.  
     
     
         12 . The method of  claim 1 , wherein the computerized list of potential matches is created by searching in a computerized list of prior customers who have made prior purchases.  
     
     
         13 . The method of  claim 12 , further including the step of identifying prior purchases made by each of the identities in the computerized list of potential matches, and further wherein the step of selecting a best match identity further comprises comparing the product purchased at the point of sale location with the prior purchases made by each of the identities in the computerized list of potential matches.  
     
     
         14 . The method of  claim 1  wherein the step of selecting a best match identity further comprising determining which identity in the list of potential matches resides closest to the point of sale location.  
     
     
         15 . The method of  claim 1 , wherein the step of selecting a best match identity further comprising determining which identity in the list of potential matches has a name that comes closest to the customer name determined from the payment mechanism.  
     
     
         16 . The method of  claim 1 , wherein the step of searching in a computerized database for identities having a name similar to the customer name further comprises 
 (1) creating a trade area for the point of sale location, the trade area being determined through an analysis of past sales at the point of sale location, the trade area defining a geographic region containing the home addresses of a first subset of shoppers that have shopped at the point of sale location, while excluding a second subset of shoppers that have shopped at the point of sale location, and    (2) searching for the identities being associated with an address within the trade area.    
     
     
         17 . A computerized system for identifying a customer in a customer database system that tracks purchase transactions, the customer having purchased a product at a point of sale location by using a payment mechanism to pay for the product purchase, the system comprising: 
 a) means for determining a customer name for the customer from the payment mechanism;    b) means for creating a computerized list of potential matches, the potential matches each being an identity of a particular person, wherein an identity constitutes information that uniquely identifies a particular person, the list of potential matches being created by 
 i) searching in a computerized database for identities having a name similar to the customer name, with similar names being determined through at least one mechanism chosen from the set including identically matching names, nicknames, abbreviations, and known variations of the customer name, and  
 ii) combining a plurality of the identities found by searching the computerized database into the computerized list of potential matches; and  
   c) means for selecting a best match identity from the computerized list of potential matches, the best match identity indicating which of the plurality of identities in the list of potential matches is to be selected as the identity for the customer name; and    d) means for using the best match identity as the identity for the customer name in the customer database system.