Enhanced online advertising system

• US 8,768,766 B2
• Filed: 03/03/2006
• Issued: 07/01/2014
• Est. Priority Date: 03/07/2005
• Status: Active Grant

• Current Assignee: Amobee, Inc. (Nexxen International Ltd.)
• Original Assignee: Turn (Nexxen International Ltd.)
• Inventors: Ellis, John R., Katiyar, Satish, Smolin, Philip Michael
• Primary Examiner(s): Bekerman, Michael
• Assistant Examiner(s): Schmucker, Michael W
• Application Number:
  US11/368,011
• Publication Number:
  US 20060212350A1
• Time in Patent Office: 3,042 Days
• Field of Search: 705/14
• US Class Current: 705/14.43
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: G06Q 30/0247 Calculate past, present or future revenues
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 30/0275 Auctions

Abstract

The system provides an automatically targeted network for text and graphical advertising based on cost-per-action bidded pricing, wherein actions comprise any of acquisitions, purchases, downloads, registrations, donations, clicks, and impressions. Contextual, search and behavioral relevance features are integrated to optimize ad selection for advertisers, who enter action objectives, associated bids, and creatives or catalog assets. The assets are automatically analyzed and stored, and ads are automatically constructed for catalog assets. When an ad request is received from a user terminal in regard to a publisher asset, e.g. a web page, the ad request is matched to a stored contextual analysis of at least a portion the publisher asset if available, and preferably to a profile associated with the user of the user terminal. The best advertisements are determined, based upon a predicted response, and are then served, i.e. displayed, at the user terminal, based upon available ad space.

Claims

1. A system implemented across a network having one or more publishing sites correspondingly associated with at least one publishing entity, each of the publishing sites comprising at least one publisher page having publishing content that is displayable to any user of the publishing site, and at least one available ad space, the system comprising:

at least one processor programmed for;

storing a plurality of anonymous profiles that are associated with a plurality of users;

analyzing the displayable publishing content accessed from the publishing sites;

storing the results of the analysis of the accessed publishing content;

receiving from one or more advertising entities across the network and storing;

one or more selectable action objectives associated with one or more advertising sites correspondingly associated with the advertising entities, and

one or more ads having selectable links from which each of the respective advertising sites can be accessed;

analyzing advertising content accessed from one or more of the advertising sites, wherein the advertising content comprises at least the ads received from the advertising entities;

receiving from a user terminal across the network an automated request for one or more ads associated with a display to a user associated with the user terminal of one of the publisher pages at the user terminal, the publisher page to be displayed comprising the displayable publishing content and one or more of the available ad spaces; and

determining if the analyzed publishing content is ready for matching to the stored ads;

wherein if the processor determines that the analyzed publishing content is ready for matching to the stored ads, the processor is programmed to

match the automated ad request to at least a portion of the analyzed publishing content for the publisher page to be displayed;

statistically match the user to one of the stored plurality of anonymous profiles, using known information associated with the user, wherein the known information comprises any of location, gender, age, interests, purchases, usage patterns, or other prior actions by the user;

predict a response to the ads by the user, wherein the prediction is at least partially based on the matched stored profile and any of the analyzed publishing content and the analyzed ads received from the advertising entities, and wherein said prediction comprises a predicted impression revenue;

determine one or more of the best stored ads based on the predicted impression revenue and an observed effective impression revenue of the stored ads, wherein the processor is programmed to track a past number of impressions and resulting actions, to estimate the observed effective impression revenue of the stored ads; and

wherein the processor is programmed to apply a blending function B to the predicted impression revenue p, and to the number of impressions i and number of resulting actions a;

B(p,i,a)

wherein for each impression i and the resulting action a, the observed effective impression revenue of the stored ads is tested for statistical significance against the predicted effective impression revenue p, wherein when the statistical significance exceeds a predetermined threshold, the processor is programmed to use the observed effective impression revenue of the stored ads as the determined effective impression revenue of the stored ads;

wherein when the statistical significance equals or is below a said predetermined threshold, then the processor is programmed to use the predicted effective impression revenue of the stored ads as the determined effective impression revenue of the stored ads; and

transmit one or more of the automatically determined best stored ads to the user terminal for integration with the publisher page to be displayed; and

wherein if the processor determines that the analyzed publishing content is not ready for matching to the stored ads, the processor is programmed to transmit an ad of a determined general relevance to the user terminal for integration with the publisher page to be displayed, wherein the processor is programmed to determine general relevance based upon any of publisher selection or system analysis.

2. The system of claim 1, wherein the processor is programmed to associate a receipt of a bid price with each of the selectable action objectives assigned by the advertiser entity.

3. The system of claim 2, wherein the processor is programmed to associate a translation of the bid prices with the action objectives to bid prices that are based on estimations of the probability that the actions will occur given a display of the associated ad to the user.

4. The system of claim 1, wherein the user terminal comprises a network enabled device comprising any of a desktop computer, a portable computer, a mobile phone, a personal digital assistant, a television, a game box and a media player.

5. The system of claim 1, wherein the processor is programmed to track completion by the user of one or more of the action objectives.

6. The system of claim 1, wherein the processor is programmed to associate selection of a bid from the advertising entity with the selected action objectives.

7. The system of claim 6, wherein the determination that the analyzed publishing content is ready for matching to the stored ads is at least partially based on a comparison of the bids associated with the selected action objectives from a plurality of advertising entities.

8. The system of claim 1, wherein the analysis of the advertising content comprises a contextual analysis of assets associated with at least one of the advertising sites, wherein the assets comprise any of ads and one or more web pages associated with corresponding advertising sites.

9. The system of claim 1, wherein the analysis of the displayable publishing content comprises a contextual analysis of at least a portion of displayable publishing content for at least one publishing site, wherein the analyzed portion of the publishing sites comprises a publisher page having at least one available ad space.

10. The system of claim 1 wherein the analysis of the displayable publishing content comprises a contextual analysis of displayable publishing content for at least one publishing page other than the publisher page having the available ad space.

11. The system of claim 1, wherein the processor is programmed for analyzing secondary data.

12. The system of claim 11, wherein the secondary data comprises information regarding any of a target audience and a locality.

13. The system of claim 1 wherein, upon selection by the user of an integrated ad displayed at the user terminal, the processor is programmed to transfer the user from the displayed publisher page to a corresponding display of an advertiser site.

14. The system of claim 13, wherein the processor is programmed to track actions at the displayed advertiser site corresponding to one or more of the action objectives selected by the advertising entity.

15. The system of claim 13, wherein the determination of at least one ad associated with the displayed advertiser site is at least partially influenced by one or more tracked actions at the displayed advertiser site.

16. The system of claim 13, wherein the processor is programmed to track by any of a beacon and a cookie located on the advertiser site.

17. The system of claim 1, wherein the selectable action objectives comprise any of acquisitions, purchases, downloads, registrations, donations, clicks, and branding impressions.

18. The system of claim 1, wherein the predicted response is based on relevance features of any of the analyzed publishing context, the analyzed advertising context and information associated with the user.

19. The system of claim 18, wherein the relevance features are represented as feature vectors, and wherein one or more of the feature vectors are mapped to a predicted probability of action.

20. The system of claim 18, wherein the relevance features comprise any of term matching, category matching, concept matching, context identification, brand power, time, action history and user behavior.

21. The system of claim 1, wherein the processor is programmed to:

receive a catalog file from at least one of the advertising entities, the catalog file comprising a plurality of asset records, wherein each of the asset records comprises a plurality of fields correspondingly associated with an asset;

analyze each of the asset records in the catalog file; and

automatically produce ads corresponding to each of the analyzed asset records.

22. A process implemented across a network having one or more publishing sites correspondingly associated with at least one publishing entity, each of the publishing sites comprising at least one publisher page having publishing content that is displayable to any user of the publishing site, and at least one available ad space, the process comprising:

providing at least one processor that is programmed to perform the steps of;

storing a plurality of anonymous profiles that are associated with a plurality of users;

analyzing the displayable publishing content;

storing the analysis of the publishing content;

receiving one or more selectable objectives associated with one or more network sites from one or more advertiser entities;

receiving a bid price correspondingly associated with each of the selectable action objectives assigned by the respective advertiser entities;

receiving and storing one or more ads associated with the network site from the respective advertiser entities;

analyzing the ads received from the advertiser entities;

receiving from a user terminal across the network an automated ad request for one or more ads associated with a display to a user of one of the publisher pages at the user terminal, the publisher page to be displayed comprising the displayable publishing content and one or more of the available ad spaces; and

determining if the analyzed publishing content is ready for matching to the stored ads;

wherein if the processor determines that the analyzed publishing content is ready for matching to the stored ads, the processor performs the steps of;

matching the automated ad request to at least a portion of analyzed publishing content for the publisher page to be displayed;

statistically matching the user to one of the stored plurality of anonymous profiles, using known information associated with the user, wherein the known information comprises any of location, gender, age, interests, purchases, usage patterns, or other prior actions by the user;

predicting a response to the ads by the user, wherein the prediction is at least partially based on the matched stored profile and any of the analyzed publishing content and the analyzed advertising ads received from the advertising entities, and wherein said prediction comprises a predicted impression revenue;

determining one or more of the best stored ads based on the predicted impression revenue and an observed effective impression revenue of the stored ads, wherein the processor is programmed to track a past number of impressions and resulting actions to estimate the observed effective impression revenue of the stored ads; and

wherein the processor is programmed to apply a blending function B to the predicted impression revenue p, and to the number of impressions i and number of resulting actions a;

B(p,i,a)

wherein for each impression i and the resulting action a, the observed effective impression revenue of the stored ads is tested for statistical significance against the predicted effective impression revenue p, wherein when the statistical significance exceeds a predetermined threshold, the processor is programmed to use the observed effective impression revenue of the stored ads as the determined effective impression revenue of the stored ads;

wherein when the statistical significance equals or is below said predetermined threshold, then the processor is programmed to use the predicted effective impression revenue of the stored ads; and

sending one or more of the determined stored ads to the user terminal or integration with the publisher page to be displayed; and

wherein the processor performs the step of transmitting an ad of a determined general relevance to the user terminal for integration with the publisher page to be displayed if the analyzed publishing content is not ready, wherein the processor performs the step of determining general relevance based upon any of publisher selection or system analysis.

23. The process of claim 22, wherein the bid price is based on any of cost per click (CPC), cost per impression (CPM), ad-based cost per action (CPA) bids, and commission-based CPA bids.

24. The process of claim 22, further comprising the step of:

automatically translating the bid prices associated with the objectives to bid prices that are based on estimations of the probability that the objectives will occur given a display of the associated ads to the user.

25. The process of claim 22, wherein the user terminal comprises a network enabled device comprising any of a desktop computer, a portable computer, a mobile phone, a personal digital assistant, a television and a media player.

26. The process of claim 22, further comprising the step of:

tracking completion by the user of one or more of the action objectives.

27. The process of claim 22, further comprising the step of:

selecting bids from the advertising entity associated with the selected action objectives.

28. The process of claim 27, wherein the determination is at least partially based on a comparison of the bids associated with the selected action objectives from a plurality of advertising entities.

29. The process of claim 22, wherein the step of analyzing the ads comprises contextually analyzing assets associated with at least one of the advertising sites, wherein the assets comprise any of ads and one or more web pages associated with corresponding advertising sites.

30. The process of claim 22, wherein the step of analyzing the displayable publishing content comprises contextually analyzing at least a portion of publishing content for at least one publishing site, wherein the analyzed portion of the publishing sites comprises a publisher page having at least one available ad space.

31. The process of claim 22, wherein the step of analyzing the displayable publishing content comprises contextually analyzing publishing content for at least one publishing page other than the publisher page having the available ad space.

32. The process of claim 22, wherein the step of analyzing the displayable publishing content further comprises analyzing secondary data.

33. The process of claim 32, wherein the secondary data comprises information regarding any of a target audience and a locality.

34. The process of claim 22, further comprising the step of:

upon selection by the user of an integrated ad displayed to the user at the user terminal, transferring the user from the displayed publisher page to a corresponding display of an advertiser site.

35. The process of claim 34, further comprising the step of:

tracking actions at the displayed advertiser site corresponding to one or more of the action objectives selected by the advertising entity.

36. The process of claim 34, wherein a ranking of at least one ad associated with the displayed advertiser site is at least partially influenced by one or more tracked actions at the displayed advertiser site.

37. The process of claim 36, wherein the tracking is provided by any of a beacon and a cookie located at the advertiser site.

38. The process of claim 22, wherein the selectable action objectives comprise any of acquisitions, purchases, downloads, registrations, donations, clicks, and branding impressions.

39. The process of claim 22, wherein the predicted response is based on relevance features of any of the analyzed publishing context, the analyzed advertising context and information associated with the user.

40. The process of claim 39, wherein the relevance features are represented as feature vectors, and wherein one or more of the feature vectors are mapped to a predicted probability of action.

41. The process of claim 39, wherein the relevance features comprise any of term matching, category matching, concept matching, context identification, brand power, time, action history and user behavior.

42. The process of claim 22, further comprising the steps of:

receiving a catalog the from at least one of the advertising entities, the catalog file comprising a plurality of asset records, wherein each of the asset records comprises a plurality of fields correspondingly associated with an asset;

automatically analyzing each of the asset records in the catalog file; and

automatically producing ads corresponding to each of the analyzed asset records.

43. A process implemented across a network having one or more publishing sites correspondingly associated with at least one publishing entity, each of the publishing sites comprising at least one publisher page having publishing content that is displayable to any user of the publishing site, and at least one available ad space, the process comprising:

providing at least one processor that is programmed to perform the steps of;

storing a plurality of anonymous profiles that are associated with a plurality of users;

analyzing the displayable publishing content;

storing the analysis of the publishing content;

receiving one or more selectable objectives associated with an advertiser campaign from an advertiser entity, the advertiser campaign comprising a plurality of assets, one of the assets comprising a network site and one or more ads;

receiving a bid price correspondingly associated with each of the selectable objectives assigned by the advertiser entity;

receiving and storing one or more of the ads associated with the advertiser campaign from the advertiser entity;

analyzing at least a portion of the ads that were received from the advertiser entity;

receiving from a user terminal across the network an automated ad request for one or more ads associated with a display to a user of one of the publisher pages at the user terminal, the publisher page to be displayed comprising the displayable publishing content and one or more of the available ad spaces; and

determining if the analyzed publishing content is ready for matching to the stored ads;

wherein if the processor determines that the analyzed publishing content is ready for matching to the stored ads, the processor performs the steps of;

matching the automated ad request to at least a portion of the contextually analyzed publishing content for the publisher page to be displayed;

statistically matching the user to one of the stored plurality of anonymous profiles, using known information associated with the user, wherein the known information comprises any of location, gender, age, interests, purchases, usage patterns, or other prior actions by the user;

predicting a response to the ads by the user, wherein the prediction is at least partially based on the matched stored profile and any of the analyzed publishing content and the analyzed advertising content of the associated ads and the analyzed ads received from the advertising entities, and wherein said prediction comprises a predicted impression revenue;

ranking one or more of the stored ads based on the predicted impression revenue and an observed effective impression revenue of the stored ads, wherein the processor is programmed to track a past number of impressions and resulting actions to estimate the observed effective impression revenue of the stored ads; and

wherein the processor is programmed to apply a blending function B to the predicted impression revenue p, and to the number of impressions i and number of resulting actions a;

B(p,i,a)

wherein for each impression i and the resulting action a, the observed effective impression revenue of the stored ads is are tested for statistical significance against the predicted effective impression revenue p, wherein when the statistical significance exceeds a predetermined threshold, the processor is programmed to use the observed effective impression revenue of the stored ads as the determined effective impression revenue of the stored ads;

if wherein when the statistical significance equals or is below a said predetermined threshold, then the processor is programmed to use the predicted effective impression revenue of the stored ads as the determined effective impression revenue of the stored ads; and

sending one or more of the ranked ads to the user terminal for integration with the publisher page to be displayed; and

wherein the processor performs the step of sending an ad having a determined general relevance to the user terminal for integration with the publisher page to be displayed if the analyzed publishing content is not ready, wherein the processor performs the step of determining general relevance based upon any of selection by a publisher associated with the publishing site, or by system analysis.

44. A process implemented across a network having one or more publishing sites correspondingly associated with at least one publishing entity, each of the publishing sites comprising at least one publisher page having publishing content that is displayable to any user of the publishing site, and at least one available ad space, the process comprising:

providing at least one processor that is programmed to perform the steps of;

storing a plurality of anonymous profiles that are associated with a plurality of users;

receiving a catalog file from an advertiser entity across the network, the catalog file comprising a plurality of asset records, wherein each of the asset records comprises a plurality of fields correspondingly associated with an asset;

receiving from the advertiser entity one or more action objectives associated with the assets;

assigning bids for the received action objectives associated with the assets, wherein the assigned bids correspond to a price corresponding to an accomplishment of a corresponding action objective;

analyzing one or more of the fields that correspond to each of the asset records;

storing the analyzed asset records;

automatically producing ads corresponding to the analyzed asset records, wherein the produced ads include the analyzed fields;

statistically matching a user at a user terminal with one of the stored plurality of anonymous profiles, wherein the matching is at least partially based upon known information about the user, wherein the known information comprises any of location, gender, age, interests, purchases, usage patterns, or other prior actions by the user;

predicting a response to the ads by the user, wherein the prediction is at least partially based on the matched stored profile and at least one of the analyzed fields associated with the automatically produced ads and any of the analyzed publishing content and the analyzed ads received from the advertising entities, and wherein said prediction comprises a predicted impression revenue;

determining one or more of the best stored ads based on the predicted impression revenue and an observed effective impression revenue of the stored ads, wherein the processor is programmed to track a past number of impressions and resulting actions and to estimate the observed effective impression revenue of the stored ads; and

wherein the processor is programmed to apply a blending function B to the predicted impression revenue p, and to the number of impressions i and number of resulting actions a;

B(p,i,a)

wherein for each impression i and the resulting action a, the observed effective impression revenue of the stored ads is are tested for statistical significance against the predicted effective impression revenue p, wherein when the statistical significance exceeds a predetermined threshold, the processor is programmed to use the observed effective impression revenue of the stored ads as the determined effective impression revenue of the stored ads;

otherwise, the processor is programmed to use the predicted effective impression revenue of the stored ads; and

sending one or more of the automatically produced ads for presentation to the user, based upon the prediction.

45. The process of claim 44, wherein the assets comprise any of products and services.

46. The process of claim 44, wherein the fields comprise any of associated SKUs, product names, product descriptions, list prices, sale prices, stock indicators, landing URLs, categories, and URLs of thumbnail images.

47. The process of claim 44, wherein the bids are assigned based on a price value of at least one of the fields.

48. The process of claim 44, wherein the bids correspond to any of cost per click (CPC), cost per impression (CPM), ad-based cost per action (CPA) bids, and commission-based CPA bids.

49. The process of claim 44, wherein the analysis comprises a contextual analysis of one or more advertiser assets.

50. The process of claim 49, wherein the analyzed advertiser assets comprise elements of a network site associated with the advertiser entity.

51. The process of claim 50, wherein the analyzed network site elements comprise one or more pages.

52. The process of claim 51, wherein at least one of the analyzed network site pages comprises a landing page.

53. The process of claim 44, further comprising the step of:

determining an ad format for one or more of the ads.

54. The process of claim 53, wherein the step of determining ad formats includes input from any of the advertiser entity, a secondary source, and an internal ad template source.

55. The process of claim 44, further comprising the step of:

storing the produced ads for use on the network.

56. The process of claim 55, wherein the processor is further programmed for:

receiving from the user terminal across the network an automated ad request for one or more ads associated with a display of one of the publisher pages at the user terminal;

wherein the step of sending one or more of the automatically produced ads is responsive to the automated ad request.

57. The process of claim 44, wherein the at least one of the analyzed fields comprises a product description.