System and method for modeling customer response using data observable from customer buying decisions

• US 20050273377A1
• Filed: 06/05/2004
• Published: 12/08/2005
• Est. Priority Date: 06/05/2004
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Pierce, Robert D., Ouimet, Kenneth J.
• Granted Patent: US 7,835,936 B2
• Time in Patent Office: Days
• US Class Current: 705/10
• 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

A computer system models customer response using observable data. The observable data includes transaction, product, price, and promotion. The computer system receives data observable from customer responses. A set of factors including customer traffic within a store, selecting a product, and quantity of selected product is defined as expected values, each in terms of a set of parameters related to customer buying decision. A likelihood function is defined for each of the set of factors. The parameters are solved using the observable data and associated likelihood function. The customer response model is time series of unit sales defined by a product combination of the expected value of customer traffic and the expected value of selecting a product and the expected value of quantity of selected product. A linear relationship is given between different products which includes a constant of proportionality that determines affinity and cannibalization relationships between the products.

Claims

1. A computer implemented method of modeling customer response, comprising:

providing data observable from customer responses;

providing an expected value of customer traffic within a store in terms of a first set of parameters;

providing an expected value of selecting a product in terms of a second set of parameters;

providing an expected value of quantity of selected product in terms of a third set of parameters;

solving for the first, second, and third sets of parameters using the data observable from customer responses; and

providing a customer response model as a combination of the expected value of customer traffic and the expected value of selecting a product and the expected value of quantity of selected product.

2. The computer implemented method of claim 1, wherein the data observable from customer responses includes transaction, product, and promotion.

3. The computer implemented method of claim 1, wherein the customer response model is a product combination of the expected value of customer traffic and the expected value of selecting a product and the expected value of quantity of selected product.

4. The computer implemented method of claim 1, wherein the expected value of customer traffic is an exponential function of the first set of parameters.

5. The computer implemented method of claim 1, further including providing a likelihood function for the customer traffic.

6. The computer implemented method of claim 5, wherein the likelihood function uses a Poisson probability distribution.

7. The computer implemented method of claim 5, wherein the first set of parameters is determined using the likelihood function and the data observable from customer responses.

8. The computer implemented method of claim 5, wherein the first set of parameters is determined through an iterative estimation process using the likelihood function and the data observable from customer responses.

9. The computer implemented method of claim 8, wherein the iterative estimation process is a maximum likelihood method.

10. The computer implemented method of claim 1, wherein the expected value of selecting a product is a function of exponential components in terms of the second set of parameters.

11. The computer implemented method of claim 10, further including providing a likelihood function for selecting a product.

12. The computer implemented method of claim 11, wherein the likelihood function uses a Binomial probability distribution.

13. The computer implemented method of claim 11, wherein the second set of parameters is determined using the likelihood function and the data observable from customer responses.

14. A method of modeling customer response based on customer buying decision, comprising:

providing expected values for a set of factors representing components of the customer buying decision, wherein each of the set of factors uses a set of parameters related to the customer buying decision;

solving for the set of parameters for each of the set of factors using data observable from customer response; and

providing a customer response model from the expected values of each of the set of factors with the solved set of parameters.

15. The method of claim 14, wherein the set of factors includes at least one from the group of customer traffic, selecting a product, and quantity of selected product.

16. The method of claim 14, wherein the set of factors includes customer traffic, selecting a product, and quantity of selected product.

17. The method of claim 16, wherein the customer response model is a product combination of the expected value of customer traffic and the expected value of selecting a product and the expected value of quantity of selected product.

18. The method of claim 14, wherein the observable data includes transaction, product, and promotion.

19. The method of claim 14, wherein one of the expected values for the set of factors is an exponential function of the set of parameters.

20. The method of claim 14, further including providing a likelihood function for each of the set of factors.

21. The method of claim 20, wherein at least one likelihood function is selected from the group of Poisson and Binomial probability functions.

22. The method of claim 20, wherein the set of parameters is evaluated using the likelihood function and the data observable from customer response.

23. The method of claim 20, wherein the set of parameters is evaluated through an iterative estimation process using the likelihood function and the data observable from customer response.

24. The method of claim 23, wherein the iterative estimation process is a maximum likelihood method.

25. A method of modeling purchasing decision processes, comprising:

providing a set of factors which define a purchasing decision process;

expressing each of the set of factors in terms of a statistical relationship involving a set of parameters related to the purchasing decision process;

solving for the set of parameters using data observable from the decision process; and

providing a model from the set of factors with the solved set of parameters.

26. The method of claim 25, wherein the data observable from the purchasing decision process includes transaction, product, and promotion.

27. The method of claim 25, wherein the set of factors includes at least one from the group of customer traffic, selecting a product, and quantity of selected product.

28. The method of claim 25, wherein the set of factors includes customer traffic, selecting a product, and quantity of selected product.

29. The method of claim 28, wherein the statistical relationship is expected values of each of the set of factors.

30. The method of claim 29, wherein the model is a product combination of the expected value of customer traffic and the expected value of selecting a product and the expected value of quantity of selected product.

31. The method of claim 29, wherein one of the expected values for the set of factors is an exponential function of the set of parameters.

32. The method of claim 25, further including providing a likelihood function for each of the set of factors.

33. The method of claim 32, wherein at least one likelihood function is selected from the group of Poisson and Binomial probability distributions.

34. The method of claim 32, wherein the set of parameters is evaluated using the likelihood function and the data observable from the decision process.

35. A method of providing a computer model of customer response, comprising:

providing a set of factors related to customer buying decision, wherein each of the set of factors uses a set of parameters related to the customer buying decision;

solving for the set of parameters using data observable from the customer buying process; and

providing a computer model of the customer response from the set of factors with the solved set of parameters.

36. The method of claim 35, wherein the data observable from the customer buying process includes transaction, product, and promotion.

37. The method of claim 35, wherein the set of factors includes customer traffic, selecting a product, and quantity of selected product.

38. The method of claim 37, wherein the computer model is a product combination of an expected value of customer traffic and an expected value of selecting a product and an expected value of quantity of selected product.

39. The method of claim 35, further including providing a likelihood function for each of the set of factors.

40. The method of claim 39, wherein the set of parameters is evaluated using the likelihood function and the data observable from the customer buying decision.

41. A computer program product usable with a programmable computer processor having a computer readable program code embodied therein, comprising:

computer readable program code which provides a set of factors related to customer buying decision, wherein each of the set of factors uses a set of parameters related to the customer buying decision;

computer readable program code which solves for the set of parameters using data observable from the customer buying process; and

computer readable program code which provides a computer model of the customer response from the set of factors with the solved set of parameters.

42. The computer program product of claim 41, wherein the set of factors includes customer traffic, selecting a product, and quantity of selected product.

43. The computer program product of claim 42, wherein the computer model is a product combination of an expected value of customer traffic and an expected value of selecting a product and an expected value of quantity of selected product.

44. The computer program product of claim 41, further including computer readable program code which provides a likelihood function for each of the set of factors.

45. The computer program product of claim 44, further including computer readable program code which solves the set of parameters using the likelihood function and the data observable from the customer buying decision.

46. A computer system for providing a model of customer response, comprising:

means for providing a set of factors related to customer buying decision, wherein each of the set of factors uses a set of parameters related to the customer buying decision;

means for solving for the set of parameters using data observable from the customer buying process; and

means for providing a computer model of the customer response from the set of factors with the solved set of parameters.

47. The computer system of claim 46, wherein the set of factors includes customer traffic, selecting a product, and quantity of selected product.

48. The computer system of claim 47, wherein the computer model is a product combination of an expected value of customer traffic and an expected value of selecting a product and an expected value of quantity of selected product.