Determining design preferences of a group

• US 20050261953A1
• Filed: 05/24/2004
• Published: 11/24/2005
• Est. Priority Date: 05/24/2004
• Status: Active Grant

• Current Assignee: Nielsen Consumer LLC (Nielsen Holdings plc)
• Original Assignee: Affinnova, Inc. (RR Donnelley & Sons Company)
• Inventors: Karty, Kevin D., Malek, Kamal M., Teller, David B.
• Granted Patent: US 7,308,418 B2
• Time in Patent Office: Days
• US Class Current: 705/10
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming
• 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 30/0203 Market surveys; Market polls

Abstract

Disclosed are methods and apparatus for conducting market research and developing product designs. The methods involve generating and presenting, typically electronically, generations of design alternatives to persons participating in the design, selection, or market research exercise. The participants transmit data indicative of their preferences among or between the presented design alternatives. Some of the data is used to conduct a conjoint analysis or non-convergent exercise to investigate the drivers of the preferences of the group or its members, and at least a portion are used to derive follow-on generations of design alternatives or proposals. The follow-on designs are preferably generated through the use of an evolutionary or genetic computer program, influenced by the participants'"'"' preferences. The process results in the generation of one or more preferred product forms and information permitting a better understanding of what attributes of the product influence the preferences of the test group members.

Claims

1. A method of analyzing the design preference tendencies of a group of selectors, the method comprising:

(a) presenting, over an electronic network, to a plurality of selectors, one or more groups of decision objects having a plurality of combinations of attributes;

(b) obtaining information from the plurality selectors indicative of a preference of the respective selectors from among the presented decision objects;

(c) using the information to evolve or to select a derived group of decision objects comprising one or more different combinations of attributes;

(d) iterating steps (a) through (c), using a derived group from step (c) to arrive at one or more preferred decision objects; and

(e) using at least some information from step (b) to implement a conjoint analysis to gather information relevant to the attribute preferences of said selectors.

2. The method of claim 1, wherein the selector comprises one or more of:

(i) a person;

(ii) a group of persons;

(iii) a proxy for a person such as a machine learning system;

(iv) a neural net, statistical or other mathematical model, or expert system;

or

(v) a combination thereof.

3. The method of claim 2 wherein a genetic algorithm is used to evolve or to select a derived group of decision objects.

4. The method of claim 2 wherein the decision objects comprise advertising material.

5. The method of claim 2 wherein the decision objects comprise packaging material.

6. The method of claim 2 wherein the decision objects comprise manufactured consumer goods.

7. The method of claim 2, further comprising the step of:

(f) generating a report describing the design preference tendencies of the group of selectors.

8. The method of claim 2, wherein step (d) further comprises iterating steps (a) through (c) for a predetermined number of iterations.

9. The method of claim 2, wherein step (d) further comprises iterating steps (a) through (c) for a predetermined length of time.

10. The method of claim 2, further comprising the step of presenting a plurality of questions to each selector for a reply.

11. The method of claim 11, wherein each selector'"'"'s reply is utilized to choose decision objects to be presented to that selector in step (a).

12. The method of claim 11, wherein more than one selector'"'"'s replies are utilized to choose decision objects to be presented to each selector in step (a).

13. The method of claim 11, wherein more than one selector'"'"'s replies are utilized to generate a report describing the design preference tendencies of the group of selectors.

14. A method of analyzing the design preference tendencies of a group of selectors, the method comprising:

(a) presenting, over an electronic network, to a plurality of selectors, one or more groups of decision objects having a plurality of combinations of attributes;

(b) obtaining data from the plurality of selectors indicative of a preference of a selector from among the presented decision objects;

(c) using at least some data from step (b) to implement a conjoint analysis to gather information relevant to the attribute preferences of the plurality of selectors;

(d) presenting, over an electronic network, to a plurality of selectors, one or more additional groups of decision objects having a plurality of combinations of attributes;

(e) obtaining information from a selector expressing a preference of that selector from among the presented decision objects;

(f) using the information to evolve or to select a derived group of decision objects comprising one or more different combinations of attributes;

(g) iterating steps (d) through (f), using a derived group from step (f) to arrive at one or more preferred decision objects; and

(h) upon achieving a stopping criterion, selecting one or a group of preferred decision objects for further development, manufacture, use, or sale.

15. The method of claim 14, wherein the selector comprises one or more of:

(i) a person;

(ii) a group of persons;

(iii) a proxy for a person such as a machine learning system;

(iv) a neural net, statistical or other mathematical model, or expert system;

or

(v) a combination thereof.

16. The method of claim 15 wherein a genetic algorithm is used to evolve or to select the derived group of decision objects.

17. The method of claim 15 wherein step (c) is performed after step (g).

18. The method of claim 15 wherein step (c) is performed after step (h).

19. The method of claim 15 wherein the results of the conjoint analysis of step (c) are used to influence the attribute combinations of the population of additional decision objects presented in step (d).

20. The method of claim 15 wherein the decision objects presented in step (a) comprise random sets of attributes.

21. The method of claim 15 wherein the decision objects presented in step (a) comprise attributes designed to enhance the efficiency of the conjoint analysis of step (c).

22. The method of claim 15 wherein before step (a), a plurality of questions is presented to each selector for a reply.

23. The method of claim 22 wherein each selector'"'"'s reply is utilized to choose decision objects to be presented to that selector.

24. The method of claim 22 wherein more than one selector'"'"'s replies are utilized to choose decision objects to be presented to each selector.

25. A method of identifying and analyzing at least one selector'"'"'s preferences for decision object attributes, the method comprising:

(a) presenting at least one decision object from a first population to the selector;

(b) obtaining data from the selector expressing that selector'"'"'s preferences for the at least one decision object;

(c) repeating steps (a) and (b) until a switching criterion is met;

(d) presenting at least one decision object from a second population to the selector;

(e) obtaining information from the selector expressing that selector'"'"'s preferences for at least one decision object;

(f) using the information to evolve at least one decision object in the second population; and

(g) repeating steps (d) through (f) until a stopping criterion is met.

26. The method of claim 25, wherein the selector comprises one or more of:

(i) a person;

(ii) a group of persons;

(iii) a proxy for a person such as a machine learning system;

(iv) a neural net, statistical or other mathematical model, or expert system;

or

(v) a combination thereof.

27. The method of claim 26 wherein step (c) further comprises, after the switching criterion is met, conducting a conjoint analysis on the obtained data.

28. The method of claim 27 wherein step (c) further comprises, selecting at least one decision object to be presented to the selector in response to the conjoint analysis.

29. The method of claim 26 wherein step (g) further comprises, after the stopping criterion is met, conducting a conjoint analysis on the obtained data and information.

30. The method of claim 26 wherein step (c) further comprises, after the switching criterion is met, utilizing the obtained data to select at least one decision object from the second population to be presented to the selector in step (d).

31. The method of claim 26, wherein a conjoint analysis is performed using either the obtained data or the obtained information.

32. The method of claim 31, wherein a conjoint analysis is performed using both the obtained data and information.

33. The method of claim 26, further comprising step:

(h) generating a report identifying each selector'"'"'s preferences for the decision object attributes.

34. The method of claim 26 wherein the switching criterion comprises a set number of repetitions of steps (a) and (b).

35. The method of claim 26, wherein the switching criterion comprises reaching a predetermined time limit.

36. The method of claim 26, wherein the stopping criterion comprises a set number or repetitions of steps (d) through (f).

37. The method of claim 26, wherein the stopping criterion comprises reaching a predetermined time limit.

38. The method of claim 26 wherein each decision object comprises advertising material.

39. The method of claim 26 wherein each decision object comprises packaging material.

40. The method of claim 26 wherein each decision object comprises manufactured consumer goods.

41. The method of claim 26 wherein, for each iteration, each selector is presented in step (d) with at least one decision object which is substantially different from that presented to any other selector.

42. A method of identifying and analyzing at least one selector'"'"'s preferences for decision object attributes, the method comprising:

(a) presenting at least one decision object from a first population to the selector;

(b) obtaining information from the selector expressing that selector'"'"'s preferences for at least one decision object;

(c) using the information to evolve at least one decision object in the first population;

(d) repeating steps (a) through (c) until a switching criterion is met;

(e) presenting at least one decision object from a second population to the selector;

(f) obtaining data from the selector expressing that selector'"'"'s preferences for at least one decision object; and

(g) repeating steps (e) and (f) until a stopping criterion is met.

43. The method of claim 42, wherein the selector comprises one or more of:

(i) a person;

(ii) a group of persons;

(iii) a proxy for a person such as a machine learning system;

(iv) a neural net, statistical or other mathematical model, or expert system;

or

(v) a combination thereof.

44. The method of claim 43 wherein step (g) further comprises, after the stopping criterion is met, conducting a conjoint analysis on the obtained data.

45. The method of claim 43 wherein step (g) further comprises, after the stopping criterion is met, conducting a conjoint analysis on the obtained data and information.

46. The method of claim 43 wherein before step (a), a plurality of questions is presented to the selector for a reply.

47. The method of claim 46 wherein the reply is utilized to choose at least one decision object from the first population to be presented to the selector.

48. The method of claim 43 wherein at least one decision object of steps (a), (b), (c), (e), and (f) comprises a partial decision object, wherein the partial decision object comprises a subset of the decision object'"'"'s attributes.

49. The method of claim 43, wherein a conjoint analysis is performed using the obtained data.

50. The method of claim 49, wherein a conjoint analysis is performed using both the obtained data and information.

51. The method of claim 50, further comprising step:

(h) generating a report identifying each selector'"'"'s preferences for the decision object attributes.

52. The method of claim 43 wherein the switching criterion comprises a set number of repetitions of steps (a) through (c).

53. The method of claim 43, wherein the switching criterion comprises reaching a predetermined time limit.

54. The method of claim 43, wherein the stopping criterion comprises a set number or repetitions of steps (e) and (f).

55. The method of claim 43, wherein the stopping criterion comprises reaching a predetermined time limit.

56. The method of claim 43 wherein each decision object comprises advertising material.

57. The method of claim 43 wherein each decision object comprises packaging material.

58. The method of claim 43 wherein each decision object comprises manufactured consumer goods.

59. The method of claim 43 wherein, for each iteration, each selector is presented in step (a) with at least one decision object which is substantially different from that presented to any other selector.

60. A method of identifying and analyzing one or more decision object attributes for which each of a plurality of selectors has a preference, the method comprising:

(a) scoring at least one attribute for each of a plurality of decision objects in response to information provided by at least one selector over an electronic network;

(b) using information derived from each attribute'"'"'s score to populate a voting window with a plurality of decision objects;

(c) iteratively presenting a plurality of decision objects to a plurality of selectors and evolving decision objects in response to preferences expressed by at least one of the plurality of selectors; and

(d) when a stopping criterion is met, returning information identifying the one or more preferred decision objects'"'"' attributes.

61. A method of identifying a subset of a larger population of decision objects for which each of a plurality of selectors has an affinity, each of the decision objects having a combination of attributes, the method comprising the steps of:

(a) presenting, over an electronic network, to each of the plurality of selectors a first group of decision objects selected from the larger set of decision objects, each decision object in the first group having a particular combination of attributes;

(b) capturing data indicative of a preference expressed for a subset of the presented decision objects by at least some of the selectors;

(c) using the captured data in a selection process to select a second group of decision objects;

(d) repeating steps (a) through (c), using the second group of step (c) as the first group of step (a), until a stopping condition is met; and

(e) performing a conjoint analysis on the captured data.