Method, algorithm, and computer program for optimizing the performance of messages including advertisements in an interactive measurable medium

• US 8,086,485 B1
• Filed: 08/14/2009
• Issued: 12/27/2011
• Est. Priority Date: 12/29/1999
• Status: Expired due to Term

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Ranka, Sanjay, Lenderman, Jason S., Weisinger, James R.
• Primary Examiner(s): Elisca, Pierre E
• Application Number:
  US12/541,528
• Time in Patent Office: 865 Days
• Field of Search: 705/10, 705/11, 705/12, 705/14, 705/27
• US Class Current: 705/10
• CPC Class Codes: G06Q 10/06315 Needs-based resource requirements planning or analysis
• CPC Class Codes: G06Q 10/0639 Performance analysis of employees; Performance analysis of enterprise or organisation operations
• 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 30/0204 Market segmentation
• CPC Class Codes: G06Q 30/0205 Location or geographical consideration
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: G06Q 30/0244 Optimization
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: G06Q 30/0277 Online advertisement
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0633 Lists, e.g. purchase orders, compilation or processing

Abstract

Method, procedure, algorithm, system, and computer program for improving and attempting to optimize the performance of messaging campaigns, particularly to marketing campaigns in which advertisements or other messages are distributed over an interactive measurable medium such as the Internet. A goal is to allocate the message alternatives to the customer population to optimize business objectives. The includes reading prior stage message state data pertaining to a prior stage including a cumulative number of trials and a cumulative number of successes for a particular message. Message performance results representing message trials and message successes from the previous stage based on the prior-stage state are then read, and a current message state is computed. A current message allocation is generated based on the current message state. Desirably the cumulative number of trials and the cumulative number of successes are discounted based on the age of the information. This inventive procedure may be applied to various message types including, for example, web site banner advertisements, electronic advertisements generally, email messages and promotions, and the like. In one embodiment, the messages are internet web site banner ads, trials are impressions of the ads presented to the user, and successes are measured by clickthroughs from the banner ad to the web site associated with the banner ad. Computer system, computer program, and computer program product associated with the inventive method and procedures are also provided.

Claims

1. A computer-implemented method, comprising:

computing, by a computer, a current message state for a context-based advertisement based on performance results for the advertisement, the performance results including a number of presentations for the context-based advertisement during one or more prior stages of a campaign and a number of actions for the context-based advertisement during the one or more prior stages of the campaign; and

generating, by a computer, allocation parameters for the context-based advertisement based on the performance results, the allocation parameters specifying a zone of a search results page to which the context-based advertisement is allocated for another stage of the campaign.

2. The method of claim 1, wherein the context-based advertisement is an advertisement targeted for presentation in a zone based on a context of the zone.

3. The method of claim 1, wherein computing a current message state comprises:

obtaining a results vector for the context-based advertisement, the results vector specifying the performance results of the context-based advertisement when presented with zones of the search results page during one or more prior stages of a message campaign; and

computing an allocation vector for the message for another stage of a message campaign based on the results vector, the allocation vector specifying the zone of the search results page to which the message is allocated during the other stage of the message campaign.

4. The method of claim 3, further comprising allocating the message to the zone based on the allocation vector.

5. The method of claim 1, wherein the number of actions comprises a number of selections of the message.

6. The method of claim 1, wherein the performance results for the message are relative to a particular zone having a context to which the context-based advertisement is targeted.

7. The method of claim 6, wherein generating allocation parameters comprises:

identifying a particular zone context for which the performance results for the message are a maximum; and

generating allocation parameters specifying that the message is allocated to zones having the particular zone context for which the performance results for the message are the maximum.

8. The method of claim 1, wherein the zone is an advertisement slot.

9. The method of claim 1, wherein the message is an online advertisement.

10. A computer-implemented method, comprising:

generating, by a computer, a current message state for each of a plurality of targeted messages, each targeted message being a message for which presentation is conditioned on targeting criteria, the current message state for each targeted message being based on a cumulative number of trials and a cumulative number of successes for the targeted message;

identifying, by a computer, a targeting criterion for each of the plurality of targeted messages; and

recommending, by a computer, a targeting criterion for a zone based on the current message states of targeted messages in the plurality of targeted messages.

11. The method of claim 10, further comprising generating a performance report for each targeting criterion based on the current message states of the plurality of targeted messages.

12. The method of claim 11, wherein the performance report provides cumulative message performance results for messages associated with each of the targeting criteria.

13. The method of claim 10, wherein the targeting criteria includes at least one targeting criterion that conditions presentation of the message based on user profile data.

14. The method of claim 10, wherein the targeting criteria includes a targeting criterion that conditions presentation of the message based on a time of day.

15. The method of claim 10, wherein recommending a targeting criterion comprises recommending the targeting criterion that is associated with a highest message performance result based on the current message states.

16. The method of claim 10, wherein each message state is based on a cumulative number of trials and a cumulative number of successes relative to a particular zone.

17. The method of claim 16, wherein the targeting criterion is recommended for the particular zone.

18. A computer-implemented method, comprising:

obtaining performance results for a plurality of targeted messages presented in a zone, the performance results including a cumulative number of previous trials and a cumulative number of previous successes for the targeted messages when presented in the zone, each targeted message being a message for which presentation is conditioned on targeting criteria;

for each of the targeted messages, generating, by a computer, a current message state based on the performance results corresponding to the message and identifying one or more targeting criteria upon which presentation of the message is conditioned; and

generating, by a computer, a performance report for targeted messages presented in the zone, the performance report specifying the targeting criteria of messages presented in the zone and relative performance measures for advertisements having the specified targeting criteria.

19. The method of claim 18, further comprising recommending advertisements for presentation in the zone based on the performance report.

20. A computer-implemented method, comprising:

computing, by a computer, a current message state for a message based on message performance results for the message, the message performance results including a cumulative number of previous trials for the message and a number of previous successes for the message; and

allocating, by a computer, the message to a zone of a search results page based on the current message state and an amount specified by an advertiser associated with the message.

21. The method of claim 20, wherein the cumulative number of previous trials and the cumulative number of previous successes for the message comprise a cumulative number of previous trials and a cumulative number of previous successes for the message during one or more prior stages of a message campaign.

22. The method of claim 21, wherein allocating the message comprises allocating the message to the zone of the search results page comprises allocating the message to the zone for a next stage of the message campaign.

23. The method of claim 20, wherein the cumulative number of previous trials comprises a cumulative number of previous impressions for the message.

24. The method of claim 20, wherein the number of previous successes comprises a cumulative number of previous selections of the message.

25. The method of claim 20, wherein allocating the message comprises:

identifying a particular zone of the search results page for which the performance results for the message are a maximum; and

allocating the message to the particular zone for which the performance results for the message are the maximum.

26. The method of claim 20, wherein the zone of the search results page is an advertisement slot on the search results page.

27. The method of claim 20, wherein the message is an online advertisement.

28. The method of claim 22, wherein computing a current message state comprises:

obtaining a results vector for the message, the results vector specifying the message performance results of the message when presented with the zone during one or more prior stages of the message campaign; and

computing an allocation vector for the message for the next stage of the message campaign based on the results vector, the allocation vector specifying one or more zones to which the message is allocated during one or more future stages of the message campaign.

29. The method of claim 28, wherein allocating the message comprises allocating the message according to the allocation vector.

30. A computer-implemented method, comprising:

computing, by a computer, a current message state for a message based on message performance results for the message, the message performance results for the message comprising a cumulative number of previous trials for the message and a cumulative number of previous successes for the message; and

providing, by a computer, configuration parameters for the message based on the current message state, the configuration parameters specifying a zone to which the message is to be allocated for one or more other trials.

31. The method of claim 30, wherein the cumulative number of previous trials and the cumulative number of previous successes for the message comprise a cumulative number of trials and a cumulative number of successes for the message during one or more prior stages of a message campaign.

32. The method of claim 31, wherein providing the configuration parameters for the message comprises providing configuration parameters that specify a zone to which the message is allocated during a next stage of the message campaign, the next stage including the one or more other trials.

33. The method of claim 32, wherein computing a current message state comprises:

obtaining, by a computer, a results vector for the message, the results vector specifying the message performance of the message when presented with the zone during one or more prior stages of the message campaign; and

computing, by a computer, an allocation vector for the message for the next stage of the message campaign based on the results vector, the allocation vector specifying one or more zones to which the message is allocated during one or more future stages of the message campaign.

34. The method of claim 33, further comprising allocating the message to the one or more zones according to the allocation vector.

35. The method of claim 30, wherein the cumulative number of previous trials comprises a cumulative number of previous impressions for the message.

36. The method of claim 30, wherein the cumulative number of successes comprises a cumulative number of previous selections of the message.

37. The method of claim 30, wherein the performance results for the message are relative to a particular zone.

38. The method of claim 37, wherein providing configuration parameters comprises:

identifying a particular zone for which the performance results for the message are a maximum; and

providing configuration parameters specifying that the message is allocated to the particular zone for which the performance results for the message are the maximum.

39. The method of claim 30, wherein a zone is an advertisement slot on a web page.

40. The method of claim 39, wherein the web page includes search results provided by a search system.

41. The method of claim 30, wherein a zone is a portion of a network.

42. The method of claim 41, wherein the portion of the network is a section of a website.

43. The method of claim 41, wherein the portion of the network is a set of related web pages.

44. The method of claim 30, wherein the message is an online advertisement.

45. A computer-implemented method, comprising:

computing, by a computer, a current message state for each of a plurality of messages, the current message state for each message being based on message performance results for the message, the message performance results for each message comprising a cumulative number of previous trials for the message and a cumulative number of previous successes for the message; and

providing, by a computer, recommended allocations of messages to zones of a network based on the current message states for the plurality of messages, the recommended allocations specifying messages for presentation in the zones.

46. The method of claim 45, wherein the cumulative number of previous trials and the cumulative number of previous successes for the message comprise a cumulative number of trials and a cumulative number of successes for the message during one or more prior stages of a message campaign.

47. The method of claim 46, wherein providing the recommended allocations of messages to zones comprises providing recommended allocations that specify messages for presentation in the zones during a next stage of the message campaign.

48. The method of claim 47, wherein computing a current message state comprises, for each of the messages:

obtaining a results vector for the message, the results vector specifying the message performance of the message when presented with the zone during one or more prior stages of the message campaign; and

computing an allocation vector for the message for the next stage of the message campaign based on the results vector, the allocation vector specifying one or more zones to which the message is allocated during one or more future stages of the message campaign.

49. The method of claim 48, further comprising allocating the messages to zones according to the allocation vectors.

50. The method of claim 45, wherein the cumulative number of previous trials comprises a cumulative number of previous impressions for the message.

51. The method of claim 45, wherein the cumulative number of successes comprises a cumulative number of previous selections of the message.

52. The method of claim 45, wherein the performance results for the message are relative to a particular zone.

53. The method of claim 52, wherein providing recommended allocations comprises, for each of the messages:

identifying a particular zone for which the performance results for the message are a maximum; and

providing a recommended allocation specifying that the message is allocated to the particular zone for which the performance results for the message are the maximum.

54. The method of claim 45, wherein the zones of the network are advertisement slots on web pages.

55. The method of claim 54, wherein one or more of the web pages includes search results provided by a search system.

56. The method of claim 45, wherein the zone of the network is a set of locations in the network.

57. The method of claim 56, wherein the set of locations is a section of a website.

58. The method of claim 56, wherein the set of locations specify a set of related web pages.

59. The method of claim 45, wherein the message is an online advertisement.

60. A computer-implemented method, comprising:

for each of a plurality of messages;

obtaining, by a computer, performance results for the message, the performance results specifying a cumulative number of previous presentations for the message and a cumulative number of previous selections of the message, and

computing, by a computer, current message state for the message based on the performance results; and

generating allocation parameters for a zone based on the current message states of the plurality of messages, the allocation parameters for the zone specifying messages allocated to the zone for one or more future presentations of the zone.

61. The method of claim 60, wherein obtaining the performance results for the message comprises obtaining cumulative performance results for the message for one or more prior stages of a message campaign.

62. The method of claim 61, wherein obtaining the cumulative performance results comprises obtaining cumulative performance results for the message relative to a zone for which allocation parameters are to be generated.

63. The method of claim 62, wherein generating allocation parameters for the zone comprises generating allocation parameters that specify messages allocated to the zone during a next stage of the message campaign, the next stage including the one or more future presentations.

64. The method of claim 63, wherein computing a current message state comprises:

obtaining a results vector for the message, the results vector specifying the message performance of the message when presented with the zone during the one or more prior stages of the message campaign; and

computing an allocation vector for the message for the next stage of the message campaign based on the results vector, the allocation vector specifying one or more messages allocated to the zone for one or more future stages of the message campaign.

65. The method of claim 64, further comprising allocating the messages to the zone according to the allocation vector.

66. The method of claim 60, wherein generating allocation parameters comprises:

identifying messages for which the performance results relative to the zone are a maximum; and

generating allocation parameters specifying that the identified messages are allocated to the zone for one or more future stages of a message campaign.

67. The method of claim 60, wherein a zone is an advertisement slot on a web page.

68. The method of claim 67, wherein the web page includes search results provided by a search system.

69. The method of claim 60, wherein a zone is a portion of a network.

70. The method of claim 69, wherein the portion of the network is a section of a website.

71. The method of claim 69, wherein the portion of the network is a set of related web pages.

72. The method of claim 60, wherein the message is an online advertisement.