Method, apparatus, and computer program product for forecasting demand using real time demand

• US 10,685,362 B2
• Filed: 06/05/2018
• Issued: 06/16/2020
• Est. Priority Date: 10/04/2012
• Status: Active Grant

• Current Assignee: Groupon Incorporated
• Original Assignee: Groupon Incorporated
• Inventors: Shariff, Shafiq, Nordquist, Derek
• Primary Examiner(s): Gills, Kurtis
• Application Number:
  US15/997,921
• Publication Number:
  US 20190012682A1
• Time in Patent Office: 742 Days
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• 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

Provided herein are systems, methods and computer readable media for managing a sales pipeline, and in some embodiments, generating demand based on real time demand and predicted demand. An example method comprises generating a virtual promotion, wherein the virtual promotion comprises a combination of a category or sub-category, a location, and a price range, calculating a probability that a particular consumer would buy the virtual offer in a predetermined time period, wherein the probability is generated at least based on historical data related to the particular consumer and one or more related consumers, determining an estimated number of units to be sold for the virtual offer as a function of at least the probability, the estimated number of units representing a predicted demand, calculating a real time demand, wherein the real time demand is generated based on a plurality of generated identification pairs for the predetermined time period, and determining, using a processor, total demand by summing the predicted demand and the real time demand.

Claims

1. A method for forecasting demand, the method comprising:

accessing, from a database, a predicted demand for at least one promotion tuple, for a specified time period, wherein the predicted demand is representative of an estimated number of units to be sold during the specified time period, and wherein the promotion tuple comprises information indicative of a category or sub-category, a location, and a price range;

calculating, via a processor, a real-time demand using data indicative of consumer activity,

wherein the calculation of the real-time demand comprises;

accessing user search data, the user search data captured, via a user interface, from an interaction, between a user device and a promotion and marketing service website or application, to identify a requested promotion, the user search data comprising at least location specific data and a category or sub-category;

generating an identification pair for the user search data, the identification pair comprising a first classification and a second classification, the first classification identifying at least a category of promotion, and the second classification identifying a location identified by the location specific data,

wherein the first classification is generated by;

normalizing the user search data, supplying the normalized user search data to a classifying model as attribute data, wherein the classifying model is a trainable classifier adapted based on a training data set of exemplary data representing exemplary terms previously determined to be semantically related to particular categories; and

distributing the real-time demand to multiple hyper-locations, multiple sub-categories, and multiple price points due to the capturing of the real-time demand being identified by or including a high level location and category or sub-category; and

determining, via the processor, a total demand, on a per category or sub-category, per location, and per price range basis by summing the predicted demand and the real time demand.

2. The method of claim 1, further comprising:

calculating the predicted demand by;

generating a virtual promotion, wherein the virtual promotion comprises a combination of a category or sub-category, a location, and a price range;

calculating a probability that a particular consumer would buy the virtual offer in a predetermined time period, wherein the probability is generated at least based on historical data related to the particular consumer and one or more related consumers;

determining an estimated number of units to be sold for the virtual offer as a function of at least the probability, the estimated number of units representing the predicted demand.

3. The method of claim 1, wherein the calculation of the real-time demand further comprises:

identifying a benchmark conversion rate;

comparing the real time demand to the benchmark conversion rate to set a modified real time demand that identifies a quantity of units that would have been purchased had the units been available;

identifying an actual purchase quantity; and

modifying the real time demand, wherein the real time demand is equal to the actual purchase quantity subtracted from the modified real time demand.

4. The method of claim 1, wherein the calculation of the real-time demand further comprises:

capturing the user search data from browsing activity and character input, wherein browsing activity includes icon selection, navigation of a hierarchal structure presented as the promotion and marketing service website or application, and location of the user device.

5. The method of claim 1, wherein the first classification is derived from captured browsing activity and text input and the second classification is derived from any of global positioning system (GPS) data indicative of a location from the user device, character input indicative of the location of the user device, or profile data associated with the user device indicative of the location of the user device.

6. The method of claim 1, wherein the distribution of the real-time demand among the multiple hyper-locations, multiple sub-categories, and multiple price points is performed in accordance with a current distribution of units among the multiple hyper-locations, the multiple sub-categories, and the multiple price points within the high level location and the category or sub-category.

7. A computer program product comprising at least one computer-readable storage medium having computer-executable program code instructions stored therein, the computer-executable program code instructions comprising program code instructions for:

accessing, from a database, a predicted demand for at least one promotion tuple, for a specified time period, wherein the predicted demand is representative of an estimated number of units to be sold during the specified time period, and wherein the promotion tuple comprises information indicative of a category or sub-category, a location, and a price range;

calculating, via a processor, a real-time demand using data indicative of consumer activity,

wherein the calculation of the real-time demand comprises;

accessing user search data, the user search data captured, via a user interface, from an interaction, between a user device and a promotion and marketing service website or application, to identify a requested promotion, the user search data comprising at least location specific data and a category or sub-category;

generating an identification pair for the user search data, the identification pair comprising a first classification and a second classification, the first classification identifying at least a category of promotion, and the second classification identifying a location identified by the location specific data,

wherein the first classification is generated by;

normalizing the user search data, supplying the normalized user search data to a classifying model as attribute data, wherein the classifying model is a trainable classifier adapted based on a training data set of exemplary data representing exemplary terms previously determined to be semantically related to particular categories; and

distributing the real-time demand to multiple hyper-locations, multiple sub-categories, and multiple price points due to the capturing of the real-time demand being identified by or including a high level location and category or sub-category; and

determining, via the processor, a total demand, on a per category or sub-category, per location, and per price range basis by summing the predicted demand and the real time demand.

8. The computer program product of claim 7, wherein the computer-executable program code instructions for the selecting of the one or more virtual offers further comprise program code instructions for:

calculating the predicted demand by;

generating a virtual promotion, wherein the virtual promotion comprises a combination of a category or sub-category, a location, and a price range;

calculating a probability that a particular consumer would buy the virtual offer in a predetermined time period, wherein the probability is generated at least based on historical data related to the particular consumer and one or more related consumers;

determining an estimated number of units to be sold for the virtual offer as a function of at least the probability, the estimated number of units representing the predicted demand.

9. The computer program product of claim 7, wherein the calculation of the real-time demand further comprises:

identifying a benchmark conversion rate;

comparing the real time demand to the benchmark conversion rate to set a modified real time demand that identifies a quantity of units that would have been purchased had the units been available;

identifying an actual purchase quantity; and

modifying the real time demand, wherein the real time demand is equal to the actual purchase quantity subtracted from the modified real time demand.

10. The computer program product of claim 7, wherein the calculation of the real-time demand further comprises:

capturing the user search data from browsing activity and character input, wherein browsing activity includes icon selection, navigation of a hierarchal structure presented as the promotion and marketing service website or application, and location of the user device.

11. The computer program product of claim 7, wherein the first classification is derived from captured browsing activity and text input and the second classification is derived from any of global positioning system (GPS) data indicative of a location from the user device, character input indicative of the location of the user device, or profile data associated with the user device indicative of the location of the user device.

12. The computer program product of claim 7, wherein the distribution of the real-time demand among the multiple hyper-locations, multiple sub-categories, and multiple price points is performed in accordance with a current distribution of units among the multiple hyper-locations, the multiple sub-categories, and the multiple price points within the high level location and the category or sub-category.

13. An apparatus comprising at least one processor and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the processor, cause the apparatus to at least:

access, from a database, a predicted demand for at least one promotion tuple, for a specified time period, wherein the predicted demand is representative of an estimated number of units to be sold during the specified time period, and wherein the promotion tuple comprises information indicative of a category or sub-category, a location, and a price range;

calculate, via a processor, a real-time demand using data indicative of consumer activity,

wherein the calculation of the real-time demand comprises;

accessing user search data, the user search data captured, via a user interface, from an interaction, between a user device and a promotion and marketing service website or application, to identify a requested promotion, the user search data comprising at least location specific data and a category or sub-category;

generating an identification pair for the user search data, the identification pair comprising a first classification and a second classification, the first classification identifying at least a category of promotion, and the second classification identifying a location identified by the location specific data,

wherein the first classification is generated by;

normalizing the user search data, supplying the normalized user search data to a classifying model as attribute data, wherein the classifying model is a trainable classifier adapted based on a training data set of exemplary data representing exemplary terms previously determined to be semantically related to particular categories; and

distributing the real-time demand to multiple hyper-locations, multiple sub-categories, and multiple price points due to the capturing of the real-time demand being identified by or including a high level location and category or sub-category; and

determine, via the processor, a total demand, on a per category or sub-category, per location, and per price range basis by summing the predicted demand and the real time demand.

14. The apparatus of claim 13, wherein the at least one memory and the computer program code are further configured to, with the processor, cause the apparatus to:

calculate the predicted demand by;

generate a virtual promotion, wherein the virtual promotion comprises a combination of a category or sub-category, a location, and a price range;

calculate a probability that a particular consumer would buy the virtual offer in a predetermined time period, wherein the probability is generated at least based on historical data related to the particular consumer and one or more related consumers;

determine an estimated number of units to be sold for the virtual offer as a function of at least the probability, the estimated number of units representing the predicted demand.

15. The apparatus of claim 13, wherein the calculation of the real-time demand further comprises:

identifying a benchmark conversion rate;

comparing the real time demand to the benchmark conversion rate to set a modified real time demand that identifies a quantity of units that would have been purchased had the units been available;

identifying an actual purchase quantity; and

modifying the real time demand, wherein the real time demand is equal to the actual purchase quantity subtracted from the modified real time demand.

16. The apparatus of claim 13, wherein the calculation of the real-time demand further comprises:

capturing the user search data from browsing activity and character input, wherein browsing activity includes icon selection, navigation of a hierarchal structure presented as the promotion and marketing service website or application, and location of the user device.

17. The apparatus of claim 13, wherein the first classification is derived from captured browsing activity and text input and the second classification is derived from any of global positioning system (GPS) data indicative of a location from the user device, character input indicative of the location of the user device, or profile data associated with the user device indicative of the location of the user device.

18. The apparatus of claim 13, wherein the distribution of the real-time demand among the multiple hyper-locations, multiple sub-categories, and multiple price points is performed in accordance with a current distribution of units among the multiple hyper-locations, the multiple sub-categories, and the multiple price points within the high level location and the category or sub-category.