Trend-factored RFM scores to improve campaign performance

• US 9,760,900 B2
• Filed: 08/08/2013
• Issued: 09/12/2017
• Est. Priority Date: 08/08/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Sewak, Mohit
• Primary Examiner(s): Boswell, Beth V
• Assistant Examiner(s): WAESCO, JOSEPH M
• Application Number:
  US13/962,371
• Publication Number:
  US 20150046218A1
• Time in Patent Office: 1,496 Days
• Field of Search: 705 731
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

A computer determines an RFM score for each of a plurality of contacts, each RFM score comprising a recency score, a frequency score and a monetary score respectively based on assigning recency values, frequency values, and monetary values of the plurality of contacts to groups. The computer generates trend factors, wherein the trend factors are based on the monetary values and timestamps of each of a plurality of transactions for each contact of the plurality of contacts. The computer generates a mean trend factor, based on the trend factors generated for each of the plurality of transactions for each contact of the plurality of contacts, and determines, by one or more processors, a trend adjusted RFM (tRFM) score based on applying the mean trend factor to at least one of the recency score, the frequency score, and the monetary score, for each contact of the plurality of contacts.

Claims

1. A computer program product for improved sorting of contacts that are grouped together based on a recency, frequency, and monetary value (RFM) score, the computer program product comprising:

one or more computer-readable storage medium(s) and program instructions stored on the one or more computer-readable storage medium(s), wherein the one or more computer-readable storage medium(s) are not transitory signals per se, the program instructions comprising;

program instructions to receive and store, transaction information for each contact of a plurality of contacts, wherein the transaction information includes;

a timestamp of each transaction of a plurality of transactions of each contact of the plurality of contacts, an identification of each contact of the plurality of contacts, a monetary value of each transaction of the plurality of transactions of each contact of the plurality of contacts, and a quantity of transactions taking place within a pre-determined period of time for each contact of the plurality of contacts;

program instructions to generate a first set of bins, a second set of bins for each first set of bins, and a third set of bins for each second set of bins, wherein a bin is a data structure configured to receive and store at least some portion of transaction information for each contact of the plurality of contacts;

program instructions to generate an RFM score for each of the plurality of contacts, wherein a respective RFM score of a contact is based on information of a sequence of consecutive transactions of the plurality of transactions for respective contacts of the plurality of contacts;

program instructions to sort a first group of contacts of the plurality of contacts to a first bin of the first set of bins, sort a second group of contacts from the first group of contacts to a second bin of the second set of bins, and sort a third group of contacts from the second group of contact to a third bin of the third set of bins, based on the RFM score generated for each respective contact of the plurality of contacts, wherein the RFM scores of contacts sorted to the third group are similar;

program instructions to generate a trend factor for the contacts of the plurality of contacts sorted to the third set of bins, based on the plurality of transactions of each contact, wherein generation of a trend factor is based on a pattern of the transaction information of the sequence of consecutive transactions of a contact of the plurality of contacts, and wherein the timestamp of each transaction of the plurality of transactions determines the sequence of consecutive transactions, and wherein the trend factor indicates a direction of change in the pattern of the transaction information of the plurality of transactions from a less recent transaction to a most recent transaction;

program instructions to generate an increasing trend factor bin and a decreasing trend factor bin;

program instructions to sort the contacts of the third group that have trend factors indicating an increase within the pattern of transaction information across the sequence of consecutive transactions of the plurality of transactions, respectively, into the increasing trend factor bin;

program instructions to sort the contacts of the third group that have trend factors that indicate a non-increasing pattern of the transaction data across the sequence of consecutive transactions of the plurality of transactions, respectively, into the decreasing trend factor bin; and

program instructions to transmit campaign communications including promotion content, directed to the contacts sorted to the increasing trend factor bin, and campaign communications including retention content, directed to the contacts sorted to the decreasing trend factor bin.

2. The computer program product of claim 1, wherein a trend factor is based on a period of time between the timestamp of a transaction and the timestamp of a previous transaction for each of the plurality of transactions of the contacts of the plurality of contacts, respectively, wherein the period of time for an initial transaction, which is the least recent transaction, of the plurality of transactions for each contact of the plurality of contacts, respectively, is assigned a value of an average of all time periods between transactions from an initial transaction to a most recent transaction for a defined period of time.

3. The computer program product of claim 1, wherein the program instructions to generate a trend factor of the contacts, comprise:

program instructions to determine a transaction and a next previous transaction for each of the plurality of transactions of the contacts, respectively; and

program instructions to generate a time interval weighted mean trend factor as a ratio of;

a summation of a product of a value of a ratio of the monetary value of a transaction and a monetary value of a next previous transaction, and a reciprocal of a duration of time between the transaction and the next previous transaction for each transaction of the plurality of transactions for the contacts, respectively,

and a summation of the reciprocal of the duration of time between the transaction and the next previous transaction for each of the plurality of transactions for the contacts, respectively, wherein the reciprocal of the duration of time for an initial transaction of the plurality of transactions for each of the contacts, respectively, is assigned a value equal to a reciprocal of an average of all time periods between transactions from the initial transaction to a most recent transaction for a defined period of time.

4. The computer program product of claim 1, wherein a trend factor of a first contact of the plurality of contacts is applied to at least one of the recency score, the frequency score, and monetary score that is most relevant, based on results of previous related transaction information.

5. The computer program product of claim 1, further comprising:

in response to a lack of historic transaction information of similar activity, program instructions to apply the trend factor of the contacts, to a median weighted component of differential weights for one of a recency value, a frequency value, and a monetary value of the RFM score of the contacts, of the plurality of contacts.

6. The computer program product of claim 1, wherein a trend factor is one of:

a median value and a mode of values of a ratio of the monetary value of a transaction and the monetary value of a next previous transaction, from the plurality of transactions of a contact of the of the plurality of contacts, respectively.

7. The computer program product of claim 1, wherein the mean trend factor of the contacts of the plurality of contacts, are determined, respectively, as a sum of a value of 1 and a best fit calculated slope of a plot of the respective trend factors generated based on transaction information of transactions from a least recent transaction to a most recent transaction of the plurality of transactions for a first contact and the plurality of transactions for a second contact, respectively, for a defined period of time.

8. A computer system for improved sorting of contacts that are grouped together based on a recency, frequency, and monetary value (RFM) score of a transaction, the computer system comprising:

one or more computer processors;

one or more computer-readable storage medium(s);

a database into which is stored transaction information that is received for each contact of a plurality of contacts, the transaction information including;

a timestamp of each transaction of a plurality of transactions of each contact of the plurality of contacts, an identification of each contact of the plurality of contacts, a monetary value of each transaction of the plurality of transactions of each contact of the plurality of contacts, and a quantity of transactions taking place within a pre-determined period of time for each contact of the plurality of contacts;

program instructions stored on one of the one or more computer-readable storage medium(s), wherein the one or more computer-readable storage medium(s) are not transitory signals per se, for execution by at least one of the one or more processors, the computer program instructions comprising;

program instructions to generate a first set of bins, a second set of bins for each first set of bins, and a third set of bins for each second set of bins, wherein a bin is a data structure configured to receive and store at least some portion of transaction information for each contact of the plurality of contacts;

program instruction to generate an RFM score for each contact of the plurality of contacts, wherein a respective RFM score of a contact is based on information of a sequence of consecutive transactions of the plurality of transactions for respective contacts of the plurality of contacts;

program instructions to sort a first group of contacts of the plurality of contacts to a first bin of the first set of bins, sort a second group of contacts from the first group of contacts to a second bin of the second set of bins, and sort a third group of contacts from the second group of contact to a third bin of the third set of bins, based on the RFM score generated for each respective contact of the plurality of contacts, wherein the RFM scores of contacts sorted to the third group are similar;

program instructions to generate a trend factor for the contacts of the plurality of contacts sorted to the third set of bins, based on the plurality of transactions of each contact, wherein generation of a trend factor is based on a pattern of the transaction information of the sequence of consecutive transactions of a contact of the plurality of contacts, and wherein the timestamp of each transaction of the plurality of transactions determines the sequence of consecutive transactions, and wherein the trend factor indicates a direction of change in the pattern of the transaction information of the plurality of transactions from a less recent transaction to a most recent transaction;

program instructions to generate an increasing trend factor bin and a decreasing trend factor bin;

program instructions to sort the contacts of the third group that have trend factors indicating an increase in the pattern of the transaction information across the sequence of consecutive transactions of the plurality of transactions, respectively, into the increasing trend factor bin;

program instructions to sort the contacts of the third group that have trend factors that indicate a non-increasing pattern of the transaction information across the sequence of consecutive transactions of the plurality of transactions, respectively, into the decreasing trend factor bin;

program instructions to transmit campaign communications including promotion content, directed to the contacts sorted to the an increasing trend factor bin, and campaign communications including retention content, directed to the contacts sorted to the decreasing trend factor bin.

9. The computer system of claim 8, wherein a trend factor is based on a period of time between the timestamp of a transaction and the timestamp of a previous transaction for each of the plurality of transactions of the contacts of the plurality of contacts, respectively, wherein the period of time for an initial transaction which is the least recent transaction, of the plurality of transactions for each contacts of the plurality of contacts, respectively, is assigned a value of an average of all time periods between transactions from an initial transaction to a most recent transaction for a defined period of time.

10. The computer system of claim 8, wherein the program instructions to determine a trend factor of the contacts of the plurality of contacts, comprises:

program instructions to determine a transaction and a next previous transaction for each of the plurality of transactions of the contacts of the plurality of contacts, respectively; and

program instructions to generate a time interval weighted mean trend factor as a ratio of;

a summation of a product of a value of a ratio of the monetary value of a transaction and a monetary value of a next previous transaction, and a reciprocal of a duration of time between the transaction and the next previous transaction for each transaction of the plurality of transactions for the contacts of the plurality of contacts, respectively,

and a summation of the reciprocal of the duration of time between the transaction and the next previous transaction for each of the plurality of transactions for the contacts of the plurality of contacts, respectively, wherein the reciprocal of the duration of time for an initial transaction of the plurality of transactions for each of the contacts of the plurality of contacts, respectively, is assigned a value equal to the reciprocal of an average of all time periods between transactions from the initial transaction to a most recent transaction for a defined period of time.

11. The computer system of claim 8, wherein the trend factor is applied to at least one of the recency score, the frequency score, and monetary score that is most relevant, based on results of previous related transaction information.

12. The computer system of claim 8, wherein the trend factor of the contacts of the plurality of contacts, are determined, respectively, as a sum of a value of 1 and a best fit calculated slope of a plot of the respective trend factors generated based on the transaction information of transactions from a least recent transaction to a most recent transaction of the plurality of transactions for the contacts of the plurality of contacts, respectively, for a defined period of time.

13. The computer system of claim 8, wherein the trend factor is one of:

a median of trend factors or a mode of trend factors.

14. The computer system of claim 8, further comprising:

in response to a lack of historic transaction information of similar activity, program instructions to apply a trend factor of the contacts of the plurality of contacts, of the plurality of contacts to a median weighted component of differential weights of a recency value, a frequency value, or a monetary value of the RFM score of the contacts of the plurality of contacts.

15. The computer system of claim 8, wherein transaction information of the contacts of the plurality of contacts is collected at a time corresponding to a completion of a transaction as designated by a timestamp of a transaction.

16. A method for improved sorting contacts that are grouped together based on a recency, frequency, and monetary value (RFM) score, the method comprising:

one or more processors receiving and storing transaction information for each contact of a plurality of contacts, wherein the transaction information includes;

a timestamp of each transaction of a plurality of transactions of each contact of the plurality of contacts, an identification of each contact of the plurality of contacts, a monetary value of each transaction of the plurality of transactions of each contact of the plurality of contacts, and a quantity of transactions taking place within a pre-determined period of time for each contact of the plurality of contacts;

one or more processors generating a first set of bins, a second set of bins for each first set of bins, and a third set of bins for each second set of bins, wherein a bin is a data structure configured to receive and store at least some portion of transaction information for each contact of the plurality of contacts;

one or more processors generating an RFM score for each of the plurality of contacts, wherein a respective RFM score of a contact is based on information of a sequence of consecutive transactions of the plurality of transactions for respective contacts of the plurality of contacts;

one or more processors sorting a first group of contacts of the plurality of contacts to a first bin of the first set of bins, sorting a second group of contacts from the first group of contacts to a second bin of the second set of bins, and sorting a third group of contacts from the second group of contact to a third bin of the third set of bins, based on the RFM score generated for each respective contact of the plurality of contacts, wherein the RFM scores of contacts sorted to the third group are similar;

one or more processors generating a trend factor for the contacts of the plurality of contacts sorted to the third set of bins, based on the plurality of transactions of each contact, wherein generation of a trend factor is based on a pattern of the transaction information of the sequence of consecutive transactions of a contact of the plurality of contacts, and wherein the timestamp of each transaction of the plurality of transactions determines the sequence of consecutive transactions, and wherein the trend factor indicates a direction of change in the pattern of the transaction information of the plurality of transactions from a less recent transaction to a most recent transaction

one or more processors generating an increasing trend factor bin and a decreasing trend factor bin;

one or more processors sorting the contacts of the third group that have a trend factor indicating an increase in the pattern of the transaction information across the sequence of consecutive transactions of the plurality of transactions, respectively, into the increasing trend factor bin;

one or more processors sorting the contacts of the that have a trend factor indicating a non-increase in the pattern of the transaction information, across the sequence of consecutive transactions of the plurality of transactions, respectively, into the decreasing trend factor bin; and

one or more processors transmitting campaign communications including promotion content, directed to the contacts sorted to the increasing trend factor bin, and campaign communications including retention content, directed to the contacts sorted to the decreasing trend factor bin.

17. The method of claim 16, wherein a trend factor is based on a period of time between the timestamp of a transaction and the timestamp of a previous transaction for each of the plurality of transactions of the contacts of the set of contacts, respectively, wherein the period of time for an initial transaction which is the least recent transaction, of the plurality of transactions for each contact of the plurality of contacts, respectively, is assigned a value of an average of all time periods between transactions from an initial transaction to a most recent transaction for a defined period of time.

18. The method of claim 16, wherein, determining a trend factor of the contacts of the set of contacts, comprises:

determining, by one or more processors, a transaction and a next previous transaction for each of a plurality of transactions for each contact of the plurality of contacts, respectively; and

generating, by one or more processors, a time interval weighted mean trend factor as a ratio of;

a summation of a product of a value of a ratio of the monetary value of a transaction and a monetary value of a next previous transaction, and a reciprocal of a duration of time between the transaction and the next previous transaction for each transaction of the plurality of transactions for each contact of the plurality of contacts, respectively,

and a summation of the reciprocal of the duration of time between the transaction and the next previous transaction for each of the plurality of transactions for a first contact and a second contact, respectively, wherein the reciprocal of the duration of time for an initial transaction of the plurality of transactions for each contact of the plurality of contacts, respectively, is assigned a value equal to a reciprocal of an average of all time periods between transactions from the initial transaction to a most recent transaction for a defined period of time.

19. The method of claim 16, wherein the trend factor of the contacts of the plurality of contacts, are applied to at least one of the recency score, the frequency score, and the monetary score that is most relevant, of the contacts of the plurality of contacts, respectively, based on results of previous related transaction information of the contacts of the plurality of contacts, respectively.

20. The method of claim 16, wherein the trend factor of a first contact and the trend factor of a second contact are determined, respectively, as a sum of a value of 1 and a best fit calculated slope of a plot of the respective trend factors generated based on transaction information of transactions from a least recent transaction to a most recent transaction of the plurality of transactions for the first contact and the plurality of transactions for the second contact, respectively, for a defined period of time.