Method, system, and computer program for determining ranges of potential purchasing amounts, indexed according to latest cycle and recency frequency, by combining re-purchasing ratios and purchasing amounts

• US 7,487,107 B2
• Filed: 06/26/2007
• Issued: 02/03/2009
• Est. Priority Date: 12/21/2001
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Blanchard, Jean-Louis, Messatfa, Hammou, Lorin, Stephane, Pavillon, Christelle
• Primary Examiner(s): Boswell; Beth V
• Assistant Examiner(s): PATS, JUSTIN
• Application Number:
  US11/768,454
• Publication Number:
  US 20070250402A1
• Time in Patent Office: 588 Days
• Field of Search: 705/10
• US Class Current: 705/7.33
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0204 Market segmentation
• CPC Class Codes: G06Q 30/0241 Advertisements

Abstract

A system and a method for selecting potential purchasers from a historical collection of confirmed purchasers. The method allows definition of a set of purchasing variables in relation to the confirmed purchasers, and computation of both a plurality of re-purchasing ratios and a plurality of purchasing amounts using the set of purchasing variables. Potential purchasing amounts are generated by combining the previous results.

Claims

1. A computer implemented method for determining ranges of purchasing amounts, said method comprising:

accessing purchasing data from a historical database of confirmed purchasers such that each confirmed purchaser has purchased at least one item having a non-null monetary value, wherein the historical database comprises the purchasing data for each confirmed purchaser within a historical duration that includes N consecutive generations and T consecutive cycles, wherein each generation is associated with a generation group consisting of all confirmed purchasers who have made a first purchase of an item of the at least one item within the associated generation, wherein each cycle of the T consecutive cycles is a time unit for analysis of the purchasing data for each confirmed purchaser, wherein N is at least 2, and wherein T is at least 3;

determining on a computer processor re-purchasing ratios from the accessed purchasing data, wherein the re-purchase ratios indicate a probability of purchasing in a next cycle of the T consecutive cycles immediately following a latest cycle of the T consecutive cycles, wherein the re-purchase ratios are indexed according to generation group, latest cycle, and recency frequency class, wherein the recency frequency class is a binary descriptor of purchases over M consecutive cycles of the T consecutive cycles, wherein the M consecutive cycles end with the latest cycle, and wherein M is at least 2;

determining purchasing amounts from the accessed purchasing data, wherein the purchasing amounts are indexed according to generation group, latest cycle, and recency frequency class;

computing on the computer processor potential purchasing amounts in the next cycle by combining the re-purchasing ratios and the purchasing amounts, wherein the potential purchasing amounts are indexed according to latest cycle and recency frequency class; and

outputting the potential purchasing amounts to an output device of a computing system;

wherein the recency frequency class is identified by a concatenation of binary values respectively corresponding to the cycles of the M consecutive cycles, wherein the binary value for the respective cycle is 1 if a purchase of one or more items of the at least one item occurred in the respective cycle, and wherein the binary value for the respective cycle is 0 if a purchase of one or more items of the at least one item did not occur in the respective cycle.

2. The method of claim 1, wherein M=2.

3. The method of claim 1, wherein M=4.

4. The method of claim 1, wherein the re-purchase ratio for each generation group is a ratio of A to B:

wherein A is the total number of confirmed purchasers in said each generation group for the next cycle pertaining to the respective latest cycle; and

wherein B is the total number of confirmed purchasers in said each generation group for the respective latest cycle and the respective recency frequency class.

5. The method of claim 1,

wherein the method further comprises computing average re-purchase ratios index according to cycle and recency frequency class by averaging the re-purchase ratios over the generation groups according to which the re-purchase ratios are indexed;

and

wherein said computer potential purchasing amounts in the next cycle comprises combining the average re-purchasing ratios and the purchasing amounts.

6. The method of claim 5,

wherein the method further comprises computing average purchasing amounts indexed according to cycle and recency frequency class by averaging the purchasing amounts over the generation groups according to which the purchasing amounts are indexed;

and

wherein said computing potential purchasing amounts in the next cycle comprises combining the average re-purchasing ratios and the average purchasing amounts.

7. The method of claim 6,

wherein the method further comprises computing minimum purchasing amounts and maximum purchasing from the purchasing amounts, such that the minimum purchasing amounts and maximum purchasing amounts are indexed according to cycle and recency frequency class;

and

wherein said computing potential purchasing amounts in the next cycle further comprises combining the average re-purchasing ratios and the minimum purchasing amounts and combining the average re-purchasing ratios and the maximum purchasing amounts.

8. The method for claim 7, wherein said computing the minimum purchasing amounts and the maximum purchasing amounts comprises computing the minimum purchasing amounts and the maximum purchasing amounts as boundaries of a 95% confidence interval with respect to the average purchasing amounts.

9. The method of claim 1, wherein the method further comprises prior to aid accessing purchasing data, collecting the purchasing data into a purchasing history table and storing the purchasing history table in the historical database in accordance with a format of the purchasing history table, and wherein said accessing purchasing data comprises accessing the purchasing data from the purchasing history table.

10. The method of claim 9, wherein the format of the purchasing history table comprises:

A column for identifying a confirmed purchaser of the confirmed purchaser;

a column for indicating a generation group to which the confirmed purchaser belongs;

a column for indicating a current cycle for the confirmed purchaser;

a column for indicating a recency frequency class associated with current cycle, and

a column for indicating a purchase amount spent by the confirmed purchaser during the current cycle.

11. The method of claim 1, wherein the output device is a display terminal or a printing device

12. A computer system comprising a processor and a computer readable memory unit coupled to the processor, said memory unit containing software commands that when executed by a processor perform the method of claim 1.

13. A computer readable medium having computer readable program code embodied therein, which upon being executed the program code performs the method of claim 1.