Computer-implemented marketing optimization systems and methods

• US 8,296,182 B2
• Filed: 08/20/2008
• Issued: 10/23/2012
• Est. Priority Date: 08/20/2008
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Chari, Manoj Keshavmurthi, Gopalakrishnan, Balaji, Liao, Yi, Opp, Michelle Gail
• Primary Examiner(s): Gart, Matthew
• Assistant Examiner(s): Danzig, Reva R
• Application Number:
  US12/194,826
• Publication Number:
  US 20100049535A1
• Time in Patent Office: 1,525 Days
• Field of Search: 705/1, 705/7, 705/14, 705/36
• US Class Current: 705/14.43
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0244 Optimization

Abstract

Computer-implemented systems and methods are provided for determining an action item from a global set of action items for a plurality of customers based on an objective function, a plurality of individual constraints, and a plurality of aggregate constraints. A plurality of offer sets is generated for each customer. An approximate highest reduced adjusted objective for each of the offer sets for each customer is calculated, and the customers are bucketed based on the highest adjusted objective value associated with each customer. The buckets are collapsed into a single bucket record containing a plurality of aggregate offer set columns, and an aggregate offer set column is selected from each bucket record for each bucket. Each bucket associated with a selected offer set is disaggregated, and the action item included in the selected offer set is stored in a computer-readable memory.

Claims

1. A computer-implemented method for determining an offer set, comprising:

accessing, using one or more data processors, a global set of action items, wherein the global set includes assignable marketing action items, and wherein action items are assignable to a plurality of customers;

generating, using the one or more data processors, initial offer sets for each customer using an objective function and one or more individual constraints, wherein an initial offer set includes one or more action items, and wherein an individual constraint limits assignable action items for a customer;

adjusting, using the one or more data processors, the objective function using one or more aggregate constraints and a Lagrange relaxation technique, wherein an aggregate constraint limits assignable action items for a plurality of customers;

calculating, using the one or more data processors, initial quality values for the initial offer sets using the adjusted objective function and a subgradient algorithm;

generating, using the one or more data processors, revised offer sets for each customer using the adjusted objective function;

calculating, using the one or more data processors, revised quality values for the revised offer sets using the adjusted objective function and the subgradient algorithm; and

determining, using the one or more data processors, an offer set for each customer using the initial quality values and the revised quality values, wherein each offer set is selected from the initial offer sets or from the revised offer sets.

2. The method of claim 1, wherein current customers and subsequent customers are associated with one or more particular individual constraints, and wherein each customer associated with the particular individual constraints is assigned one of a pool of offer sets associated with the particular individual constraints.

3. The method of claim 1, wherein the Lagrange relaxation technique includes a linear programming relaxation technique.

4. The method of claim 1, wherein the Lagrange relaxation technique uses an OCTANE heuristic.

5. The method of claim 1, further comprising:

determining an estimated value for each of the plurality of customers; and

assigning the plurality of customers into buckets based upon the estimated values, wherein a particular offer set for a customer is generated based upon a bucket assigned to the customer.

6. The method of claim 5, wherein the particular offer set for the bucket is selected using an integer programming heuristic.

7. The method of claim 1, wherein determining the offer sets for each customer includes using a gradient constraint fixing heuristic.

8. The method of claim 1, wherein the aggregate constraints include an agent constraint.

9. The method of claim 8, wherein the agent constraint identifies a maximum number of customers that an agent is permitted to contact.

10. A system for determining an offer set, comprising:

one or more processors;

one or more computer-readable storage mediums containing instructions configured to cause the one or more processors to perform operations including;

accessing a global set of action items, wherein the global set includes assignable marketing action items, and wherein action items are assignable to a plurality of customers;

generating initial offer sets for each customer using an objective function and one or more individual constraints, wherein an initial offer set includes one or more action items, and wherein an individual constraint limits assignable action items for a customer;

adjusting the objective function using one or more aggregate constraints and a Lagrange relaxation technique, wherein an aggregate constraint limits assignable action items for a plurality of customers;

calculating initial quality values for the initial offer sets using the adjusted objective function and a subgradient algorithm;

generating revised offer sets for each customer using the adjusted objective function;

calculating revised quality values for the revised offer sets using the adjusted objective function and the subgradient algorithm; and

determining an offer set for each customer using the initial quality values and the revised quality values, wherein each offer set is selected from the initial offer sets or from the revised offer sets.

11. The system of claim 10, wherein current customers and subsequent customers are associated with one or more particular individual constraints, and wherein each customer associated with the particular individual constraints is assigned one of a pool of offer sets associated with the particular individual constraints.

12. The system of claim 10, wherein the Lagrange relaxation technique includes a linear programming relaxation technique.

13. The system of claim 10, wherein the Lagrange relaxation technique uses an OCTANE heuristic.

14. The system of claim 10, wherein the operations further include:

determining an estimated value for each of the plurality of customers; and

assigning the plurality of customers into buckets based on the estimated values, wherein a particular offer set for a customer is generated based on a bucket assigned to the customer.

15. The system of claim 14, wherein the particular offer set for the bucket is selected using an integer programming heuristic.

16. The system of claim 10, wherein determining the offer sets for each customer uses a gradient constraint fixing heuristic.

17. The system of claim 10, wherein the linking constraints further include an agent constraint.

18. The system of claim 17, wherein the agent constraint identifies a maximum number of customers that an agent is permitted to contact.

19. A computer-program product for determining an offer set, tangibly embodied in a machine-readable storage medium, including instructions configured to cause a data processing apparatus to:

access a global set of action items, wherein the global set includes assignable marketing action items, and wherein action items are assignable to a plurality of customers;

generate initial offer sets for each customer using an objective function and one or more individual constraints, wherein an initial offer set includes one or more action items, and wherein an individual constraint limits assignable action items for a customer;

adjust the objective function using one or more aggregate constraints and a Lagrange relaxation technique, wherein an aggregate constraint limits assignable action items for a plurality of customers;

calculate initial quality values for the initial offer sets using the adjusted objective function and a subgradient algorithm;

generate revised offer sets for each customer using the adjusted objective function;

calculate revised quality values for the revised offer sets using the adjusted objective function and the subgradient algorithm; and

determine an offer set for each customer using the initial quality values and the revised quality values, wherein each offer set is selected from the initial offer sets or from the revised offer sets.