System and method for measuring advertising effectiveness

• US 10,282,748 B2
• Filed: 09/15/2014
• Issued: 05/07/2019
• Est. Priority Date: 02/20/2013
• Status: Active Grant

• Current Assignee: DataLogix Holdings, Inc. (Oracle Corporation)
• Original Assignee: DataLogix Holdings, Inc. (Oracle Corporation)
• Inventors: Opie, Robin, Thompson, Audrey Lynn, Wandler, Damian Vallejo, Hussen, Donald Edward
• Primary Examiner(s): Long, Fonya M
• Assistant Examiner(s): Nguyen, Thuy N
• Application Number:
  US14/486,172
• Publication Number:
  US 20150006282A1
• Time in Patent Office: 1,695 Days
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: G06Q 30/0246 Traffic

Abstract

In some embodiments, a method defines a test group from a population of consumers that have been exposed to a communication associated with a promoted entity. The test group includes a (1) retailer profile, and (2) a first purchase amount associated with the promoted entity. The method also includes receiving purchase information associated with a third party. The purchase information includes (1) a second purchase amount associated with the promoted entity, and (2) a third purchase amount associated with the promoted entity. The method further includes defining a first multiplier based on a relationship between the first purchase amount and the second purchase amount, and defining a second multiplier based on a relationship between the first purchase amount and the third purchase amount. The method additionally includes sending a signal indicative of a combined multiplier based on (1) the first multiplier, and (2) the second multiplier.

Claims

1. A system comprising:

one or more hardware processors;

one or more computer readable media that storing a sequence of instruction that, when executed by the one or more hardware processors cause;

collecting, by a collection module from a first set of one or more servers, a set of campaign data, associated with an online campaign, wherein the campaign data maps household information for a plurality of households to digital activity metrics;

populating, by the collection module, a set of records in a database with information extracted from the campaign data, each respective record in the set of records storing information for a different respective household of the plurality of households including one or more digital activity metrics associated with the respective household, wherein the database is accessible to a matching module and the collection module removes personal identifiable information from the campaign data prior to populating the database with information extracted from the campaign data;

for each respective record in the database, generating, by the collection module, a respective anonymous identifier by hashing an attribute string comprising a plurality of attributes extracted from the campaign data and associated with a respective household of the plurality of households;

encrypting each anonymous identifier in the database with an encryption key;

collecting, by the collection module from a second set of one or more servers, a set of transaction data that maps household information for a particular household of the plurality of households to one or more offline activity metrics;

hashing, by the collection module, an attribute string extracted from the transaction data and encrypting the hashed attribute string with the encryption key used to encrypt the anonymous identifiers;

matching, by the matching module, the transaction data to a record in the set of records by comparing the hashed attribute string encrypted by the encryption key to anonymous identifiers generated by the collection module and encrypted by the encryption key, wherein the matching module is prevented from directly interacting with the first set of one or more servers and the second set of one or more servers;

in response to matching transaction data to the record in the set of records, generating a combined record in the database that links one or more digital activity metrics to one or more offline activity metrics for an anonymous household represented by the anonymous identifier.

2. The system of claim 1, further comprising a measurement module that generates a score based on at least one of a household income or a digital activity measurement.

3. The system of claim 1, further comprising a measurement module configured to define a score for each household from the plurality of households based on at least one of a number of digitally active consumers in a household, amount of time spent online, or number of categories of websites browsed.

4. The system of claim 1, further comprising a measurement module configured to:

partition, using a logistic regression model, the set of records in the database into a plurality of clusters, including a first cluster and a second cluster, wherein the first cluster includes records for households from the plurality of households that (1) were exposed to a communication for a promoted entity, and (2) were assigned a score within a range of scores, wherein the second cluster includes records for households from the plurality of households that (1) were not exposed to the communication for the promoted entity, and (2) were assigned a score within the range of scores;

determine an initial effectiveness value for the online campaign based on comparing data associated with the first cluster of records with data associated with the second cluster of records in the database;

determine a combined multiplier based on a relationship between the first cluster and the second cluster; and

determine a final effectiveness value for the online campaign by multiplying the initial effectiveness value by the combined multiplier.

5. The system of claim 4, wherein the measurement module is further configured to send a signal indicative of the final effectiveness value obtained through the measurement module by multiplying the initial effectiveness value by the combined multiplier.

6. The system of claim 1, further comprising a visualization module configured to generate different views of household data stored in the combined record.

7. The system of claim 1, further comprising a measurement module configured to:

partition, in the database based on a set of alignment characteristics, the set of records into a set of clusters;

partition each cluster in the set of clusters into a plurality of sub-groups using a logistic regression model;

performing one or more diagnostic modifications by removing outlier records from at least one cluster in the set of clusters.

8. A method, comprising:

collecting, by a collection module from a first set of one or more servers, a set of campaign data, associated with an online campaign, wherein the campaign data maps household information for a plurality of households to digital activity metrics;

populating, by the collection module, a set of records in a database with information extracted from the campaign data, each respective record in the set of records storing information for a different respective household of the plurality of households including one or more digital activity metrics associated with the respective household, wherein the database is accessible to a matching module and the collection module removes personal identifiable information from the campaign data prior to populating the database with information extracted from the campaign data;

for each respective record in the database, generating, by the collection module, a respective anonymous identifier by hashing an attribute string comprising a plurality of attributes extracted from the campaign data and associated with a respective household of the plurality of households;

encrypting each anonymous identifier in the database with an encryption key;

collecting, by the collection module from a second set of one or more servers, a set of transaction data that maps household information for a particular household of the plurality of households to one or more offline activity metrics;

hashing, by the collection module, an attribute string extracted from the transaction data and encrypting the hashed attribute string with the encryption key used to encrypt the anonymous identifiers;

matching, by the matching module, the transaction data to a record in the set of records by comparing the hashed attribute string encrypted by the encryption key to anonymous identifiers generated by the collection module and encrypted by the encryption key, wherein the matching module is prevented from directly interacting with the first set of one or more servers and the second set of one or more servers;

in response to matching transaction data to the record in the set of records, generating a combined record in the database that links one or more digital activity metrics to one or more offline activity metrics for an anonymous household represented by the anonymous identifier.

9. The method of claim 8, further comprising generating, by a measurement module, a score based on at least one of a number of digitally active consumers in a household, amount of time spent online, or number of categories of websites browsed.

10. The method of claim 8, further comprising defining a score for each household from the plurality of households based on at least one of a number of digitally active consumers in a household, amount of time spent online, or number of categories of web sites browsed.

11. The method of claim 8, further comprising:

partition, by a measurement model using a logistic regression model, the set of records in the database into a plurality of clusters, including a first cluster and a second cluster, wherein the first cluster includes records for households from the plurality of households that (1) were exposed to a communication for a promoted entity, and (2) were assigned a score within a range of scores, wherein the second cluster includes records for households from the plurality of households that (1) were not exposed to the communication for the promoted entity, and (2) were assigned a score within the range of scores;

determining, by the measurement module, an initial effectiveness value for the online campaign based on comparing data associated with the first cluster of records with data associated with the second cluster of records in the database;

determining, by a measurement module, a combined multiplier based on a relationship between the first cluster and the second cluster; and

determining, by a measurement module a final effectiveness value for the online campaign by multiplying the initial effectiveness value by the combined multiplier.

12. The method of claim 11, sending a signal indicative of the final effectiveness value obtained through the measurement module by multiplying the initial effectiveness value by the combined multiplier.

13. The method of claim 8, further comprising generating, by a visualization module, different views of household data stored in the combined record.

14. A non-transitory processor-readable medium storing code representing instructions to cause a processor to perform a process, the code comprising code to:

collect, by a collection module from a first set of one or more servers, a set of campaign data, associated with an online campaign, wherein the campaign data maps household information for a plurality of households to digital activity metrics;

populate, by the collection module, a set of records in a database with information extracted from the campaign data, each respective record in the set of records storing information for a different respective household of the plurality of households including one or more digital activity metrics associated with the respective household, wherein the database is accessible to a matching module and the collection module removes personal identifiable information from the campaign data prior to populating the database with information extracted from the campaign data;

for each respective record in the database, generate, by the collection module, a respective anonymous identifier by hashing an attribute string comprising a plurality of attributes extracted from the campaign data and associated with a respective household of the plurality of households;

encrypt each anonymous identifier in the database with an encryption key;

collect, by the collection module from a second set of one or more servers, a set of transaction data that maps household information for a particular household of the plurality of households to one or more offline activity metrics;

hash, by the collection module, an attribute string extracted from the transaction data and encrypt the hashed attribute string with the encryption key used to encrypt the anonymous identifiers;

match, by the matching module, the transaction data to a record in the set of records by comparing the hashed attribute string encrypted by the encryption key to anonymous identifiers generated by the collection module and encrypted by the encryption key, wherein the matching module is prevented from directly interacting with the first set of one or more servers and the second set of one or more servers;

in response to matching transaction data to the record in the set of records, generating a combined record in the database that links one or more digital activity metrics to one or more offline activity metrics for an anonymous household represented by the anonymous identifier.

15. The non-transitory processor-readable medium storing code representing instructions to cause a processor to perform a process of claim 14, the process further comprising code to generate, by a measurement module, a score based on at least one of a number of digitally active consumers in a household, amount of time spent online, or number of categories of websites browsed.

16. The non-transitory processor-readable medium storing code representing instructions to cause a processor to perform a process of claim 14, the process further comprising code to generate, by a measurement module, a score for each household from the plurality of households based on at least one of a number of digitally active consumers in a household, amount of time spent online, or number of categories of websites browsed.

17. The non-transitory processor-readable medium storing code representing instructions to cause a processor to perform a process of claim 14, the process further comprising code to:

partition, using a logistic regression model, the set of records in the database into a plurality of clusters, including a first cluster and a second cluster, wherein the first cluster includes records for households from the plurality of households that (1) were exposed to a communication for a promoted entity, and (2) were assigned a score within a range of scores, wherein the second cluster includes records for households from the plurality of households that (1) were not exposed to the communication for the promoted entity, and (2) were assigned a score within the range of scores;

determine, by a measurement module, an initial effectiveness value for the online campaign based on comparing data associated with the first cluster of records with data associated with the second cluster of records in the database;

determine, by a measurement module, a combined multiplier based on a relationship between the first cluster and the second cluster; and

determine, by a measurement module a final effectiveness value for the online campaign by multiplying the initial effectiveness value by the combined multiplier.

18. The non-transitory processor-readable medium storing code representing instructions to cause a processor to perform a process of claim 17, the process further comprising code to:

send a signal indicative of the final effectiveness value obtained through the measurement module by multiplying the initial effectiveness value by the combined multiplier.

19. The non-transitory processor-readable medium storing code representing instructions to cause a processor to perform a process of claim 14, the process further comprising code to:

generate, by a visualization module, different views of household data stored in the combined record.