Dynamic promotion analytics

Patent metadata

Field	Value
Publication number	US-11263659-B2
Application number	US-201916427380-A
Country	US
Kind code	B2
Filing date	May 31, 2019
Priority date	May 8, 2012
Publication date	Mar 1, 2022
Grant date	Mar 1, 2022

Abstract

A promotion program analytical system and method is disclosed. The promotion program analytical system and method selects a promotion program to offer to a consumer. Selection of the promotion program to present to the consumer includes determining a probability that the consumer will accept the promotion program. The probability of acceptance may be determined based on past performance data of similar promotion programs, and also past performance data on the promotion program itself when it is available.

First claim

What is claimed is:

1. A method for determining whether to provide a particular offer, from a promotion program, to a consumer device associated with a particular consumer, via an electronic communication, the method comprising:
receiving, from the consumer device associated with the particular consumer over a network, consumer attribute data associated with the particular consumer;
deriving, via a processor, from the promotion program, a first promotion program attribute value of a first promotion program attribute and a duration, wherein the duration is comprised of a plurality of time periods;
generating, using a historical predictive model, a historical predicted probability that the particular consumer will accept the particular offer from the promotion program, wherein the historical predictive model is configured to output the historical predicted probability based on performance data from one or more different promotion programs and the consumer attribute data associated with the particular consumer,
wherein generating the historical predicted probability comprises:
correlating the first promotion program attribute value to historical predicted acceptances of offers derived from performance data of offers from the one or more different promotion programs;

generating, using a promotion program predictive model, a promotion program predicted probability that the particular consumer will accept the particular offer, wherein the promotion program predictive model is configured to output the promotion program predicted probability based on performance data from the promotion program and the consumer attribute data associated with the particular consumer,
wherein generating the promotion program predicted probability comprises:
determining a first promotion program predicted probability value, the first promotion program predicted probability value correlated to previous promotions from the promotion program having values of the first promotion program attribute that satisfy the first promotion program attribute value;

determining whether an amount of performance data is greater than a minimum threshold amount;
in an instance in which the determination is made that the amount of performance data is not greater than the minimum threshold, calculating a final predicted probability that the particular consumer will accept the particular offer to be a function of only the historical predicted probability while disregarding the promotion program predicted probability;
calculating whether the amount of performance data is greater than a second minimum threshold in an instance in which the amount of performance data is greater than the minimum threshold;
in an instance in which the amount of performance data is greater than the minimum threshold and is greater than the second minimum threshold, calculating the final predicted probability that the particular consumer will accept the particular offer to be a function of only the promotion program predicted probability while disregarding the historical predicted probability;
in an instance in which the amount of performance data is greater than the minimum threshold and is not greater than the second minimum threshold, calculating the final predicted probability that the particular consumer will accept the particular offer to be a function of both the promotion program predicted probability and the historical predicted probability;
dynamically and iteratively updating the promotion program predictive model as additional real performance data from the promotion program is available, wherein the additional real performance data includes performance data from a previous time period;
as a result of the updating of the promotion program predictive model, re-determining sufficiency of the performance data and re-calculating the final predicted probability that the particular consumer will accept the particular offer, resulting in gradual migration from a reliance on the output of the historical predicted probability based on performance data from one or more different promotion programs to reliance on output of the promotion program predicted probability based on performance data from the promotion program; and
providing, for display on the consumer device, the particular offer via the electronic communication, only upon a determination that the final predicted probability that the particular consumer will accept the particular offer is within a predefined delta based on at least the consumer attribute data associated with the particular consumer.

2. The method of claim 1 , further comprising:
determining that the final predicted probability that the particular consumer will accept the particular offer is not within the predefined delta; and
determining whether the performance data indicates that a desired number of consumers in a selected grouping have been provided the particular offer, wherein the performance data is indicative of a number of consumers with the derived first promotion program attribute value that received the particular offer from the promotion program,
wherein determining whether the performance data is insufficient comprises comparing the number of consumers with the derived first promotion program attribute value that received the particular offer from the promotion program to a predetermined number.

3. The method of claim 1 , wherein the performance data includes a number of consumers that accepted the particular offer from the promotion program in a previous time period or a percentage of consumers that accepted the particular offer from the promotion program.

4. The method of claim 2 , further comprising:
determining whether there are any other consumers to consider in the selected grouping that have not yet been provided the particular offer from the promotion program.

5. The method of claim 4 , further comprising:
in an instance in which the determination is made that there are not other consumers to consider in the selected grouping that have not yet been provided the particular offer from the promotion program, increasing the delta.

6. The method of claim 4 , further comprising:
in an instance in which the determination is made that there are not other consumers to consider in the selected grouping that have not yet been provided the particular offer from the promotion program; and
calculating a new final predicted probability that a new particular consumer will accept the particular offer; and
determining that the new final predicted probability that the new particular consumer will accept the particular offer is within the predefined delta; and
providing, for display on a consumer device associated with the new particular consumer, the particular offer via the electronic communication.

7. The method of claim 1 , further comprising:
dynamically updating the promotion program predictive model at multiple discrete times during the promotion program duration by at least one of:
updating the promotion program predictive model to account for first period offers outstanding during the first period and accepted during a second period; and updating the promotion program predictive model to account for second period offers accepted during the second offer period.

8. A computer program product for determining whether to provide a particular offer, from a promotion program, to a consumer device associated with a particular consumer, via an electronic, the computer program product, stored on a non-transitory computer readable medium, comprising instructions that when executed on one or more computers cause the one or more computers to perform operations the operations comprising:
receiving, from the consumer device associated with the particular consumer over a network, cons

Assignees

• Groupon Inc

Inventors

• Chang Kevin
• Thacker David
• Gauthier Paul
• Aggarwal Amit

Classifications

• G06Q30/0242 Primary

  Determining effectiveness of advertisements · CPC title