System and method for optimizing search result listings

• US 8,341,017 B2
• Filed: 03/22/2004
• Issued: 12/25/2012
• Est. Priority Date: 01/09/2004
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Payne, Christopher D., Watson, Eric B., Salas, Sally K., Krishnan, Sreeram
• Primary Examiner(s): Weiss, John G.
• Assistant Examiner(s): POUNCIL, DARNELL A
• Application Number:
  US10/805,873
• Publication Number:
  US 20050154718A1
• Time in Patent Office: 3,200 Days
• Field of Search: 705/14
• US Class Current: 705/14.43
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 7/00 Methods or arrangements for processing data by operating upon the order or content of the data handled logic circuits H03K19/00
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A method, system, and computer-accessible medium are provided for optimizing search result listings in a search engine user interface. The system and method provide an automated way of promoting and demoting the placement of a search result listing between more or less desirable locations, based on the listing'"'"'s performance in a given placement. Promoting and demoting placements is based on a threshold performance for the listing against which the actual performance of a listing is measured. The threshold performance may be advantageously varied to accommodate different revenue and relevance goals for a particular search engine or portion of the search engine'"'"'s market.

Claims

1. A computer-implemented method performed on a first computer for displaying on a screen coupled to a second computer, an optimized placement of search result listings displayed in a search Web page, the method comprising:

said first computer measuring an actual performance of each listing in a set of listings located in one or more Web pages containing a search term and a search result corresponding to the search term, wherein the actual performance comprises at least one of click-through rate (CTR) and context of the listing, and wherein the set of listings includes a plurality of listings;

assigning an expected performance for each listing in the set of listings based on their respective locations, wherein the expected performance comprises a user defined expected CTR and the context of the listing;

determining an average expected performance for the set of listings;

tiering the average expected performance of the set of listings based at least on the number of listings in the set of listings to produce a tiered expected performance for the set of listings, wherein the average expected performance is increased when the set of listings includes more than a predefined number of listings and the average expected performance is decreased when the set of listing includes less than a predefined number of listings;

determining an overall average performance for the set of listings based on the actual performance for each listing in the set of listings;

promoting the set of listings to a more prominent location when the overall average performance of the set of listings indicates that overall the actual performance of the listings in the set of listings is better than the tiered expected performance of the set of listings;

demoting the set of listings to a less prominent location when the overall average performance indicates that overall the actual performance of the set of listings is poorer than the tiered expected performance of the set of listings; and

displaying a listing from the set of listings at the more prominent location or at the less prominent location.

2. The method of claim 1, wherein assigning the expected performance includes increasing the expected performance when the listing in the set of listings is in a more prominent location, and decreasing the expected performance when the listing is in a less prominent location.

3. The method of claim 2, wherein the listing in the set of listings is in a more prominent location when it is included in the search result.

4. The method of claim 2, wherein the listing in the set of listings is in a more prominent location when it is located near the search term.

5. The method of claim 2, wherein the listing in the set of listings is in a less prominent location when it is located in a sidebar on the Web page.

6. The method of claim 2, wherein the listing in the set of listings is in a less prominent location when it is located separately from the search result.

7. The method of claim 1, further comprising increasing the expected performance of the listing in the set of listings based on a context of the listing, wherein the context of the listing comprises factors that increase performance, including at least one of a position of the listing above other listings, a larger size of the listing relative to other listings, a distinctive formatting applied to the display of the listing, including a contrasting color, a highlighting, an animation, and a graphic, and a good performance of a neighboring listing.

8. The method of claim 1, further comprising decreasing the expected performance of the listing in the set of listings based on a context of the listing, wherein the context of the listing comprises factors that decrease performance, including at least one of a position of the listing below other listings, a smaller size of the listing relative to other listings, a lack of distinctive formatting applied to the display of the listing, and a poor performance of a neighboring listing.

9. The method of claim 1, wherein the performance of a listing is a click-through rate, where the click-through rate is derived from a number of times the listing is displayed in the Web page as compared to a number of times the listing is clicked after being displayed.

10. The method of claim 1, wherein the listing in the set of listings is an unpaid listing, and the expected performance is a threshold level that is tuned to optimize unpaid listing relevance.

11. The method of claim 1, wherein the listing in the set of listings is a paid listing that generates advertising revenue each time it is clicked, and the expected performance is a threshold level that is tuned to optimize paid listing revenue.

12. The method of claim 1, wherein the expected performance of a listing is a threshold performance level that is tuned to a particular market.

13. The method of claim 1, wherein the overall average performance of the set of listings is based on the expected performance of each listing in the set, wherein the expected performance varies based on a location in a Web page of each listing in the set.

14. The method of claim 1, wherein the actual performance is better than the tiered expected performance when the actual performance exceeds the tiered expected performance, and poorer when the actual performance falls short of the tiered expected performance.

15. One or more computer-readable media having computer-executable instructions embodied thereon that, when executed perform a search results optimization system comprising:

a performance measurement process that measures an actual performance of a listing appearing in a search results Web page against an expected performance level, the expected performance level including settings that are tunable to accommodate different performance expectations in one or more search engine markets, wherein the actual performance comprises at least one of click-through rate (CTR) and context of the listing, wherein the context of the listing does not include performance data of another listing, wherein the expected performance level comprises a user defined expected CTR that is adjusted based on whether the listing appears in a more prominent or less prominent location;

a listing placement process that determines that a listing should be promoted to a more prominent location selected from a group of at least one more prominent location when the actual performance measures higher than the expected performance level, and that determines that the listing should be demoted demotes the listing to a less prominent location selected from a group of at least one less prominent location when the actual performance measures lower than the expected performance level; and

a displaying process that displays the listing at the more prominent location or at the less prominent location.

16. The computer-readable media of claim 15, wherein to measure an actual performance includes capturing a number of impressions of a listing and a number of clicks on a listing, and a current location of the listing relative to a location of the search results on the Web page.

17. The computer-readable media of claim 15, wherein the more prominent location is a location in which a listing is expected to receive a greater number of clicks than that received in the less prominent location.

18. The computer-readable media of claim 17, wherein the listing is in the more prominent location when it is located near the search term.

19. The computer-readable media of claim 17, wherein the listing is in the less prominent location when it is located in a sidebar on the Web page.

20. The computer-readable media of claim 17, wherein the listing is in the less prominent location when it is located separately from the search result.

21. The computer-readable media of claim 15, wherein the expected performance levels in a performance settings repository are tuned to optimize at least one of advertising revenue and search result relevance.

22. The computer-readable media of claim 15, wherein the performance measurement process takes a context of the listing into consideration when measuring the actual performance of the listing against the expected performance level for the listing.

23. The computer-readable media of claim 15, wherein the context of the listing comprises factors that increase performance, including at least one of a position of the listing above other listings, a larger size of the listing relative to other listings, a distinctive formatting applied to the display of the listing, including a contrasting color, a highlighting, an animation, and a graphic, and a good performance of a neighboring listing, and factors that decrease performance, including at least one of a position of the listing below other listings, a smaller size of the listing relative to other listings, a lack of distinctive formatting applied to the display of the listing, and a poor performance of a neighboring listing.

24. The computer-readable media of claim 15, wherein the actual performance measures higher than the expected performance level when the actual performance is greater than the expected performance, and measures lower when the actual performance is less than the expected performance.

25. One or more computer-readable media having computer-executable instructions embodied thereon that, when executed perform a method for facilitating the optimal placement of search result listings in a search result user interface, the method comprising:

placing a listing for a search result in an initial location based on an expected click-through rate (CTR);

capturing an actual CTR of the listing, wherein capturing the actual CTR of the listing includes capturing a location of the listing and a context of the listing when it was clicked;

normalizing the actual CTR based on the location, wherein normalizing the actual CTR adjusts the actual CTR down when the location has a positive influence on the actual CTR, and adjusts the actual CTR up when the location has a negative influence on the actual CTR;

normalizing the actual CTR based on the context, wherein normalizing the actual CTR adjusts the actual CTR down when the context has a positive influence on the actual CTR, and adjusts the actual CTR up when the context has a negative influence on the actual CTR; and

promoting the listing to a more desirable location when the normalized CTR is better than the expected CTR, and demoting the listing to a less desirable location when the normalized CTR is worse than the expected CTR, wherein promoting and demoting the listing is independent of the promotion or demotion of any other listings.

26. The computer-readable media of claim 25, wherein the context of the listing comprises factors that influence performance, including at least one of a position of the listing relative to other listings appearing in the search results user interface, a size of the listing relative to other listings, a presence or absence of distinctive formatting applied to the display of the listing relative to the other listings, including a contrasting color, a highlighting, an animation, and a graphic, and a CTR of an adjacent listing.

27. The computer-readable media of claim 25, wherein the listing is a pay-per-click listing and the expected CTR is set to optimize revenue earned from the listing.

28. The computer-readable media of claim 25, wherein the listing is an unpaid listing and the expected CTR is set to optimize relevance of the placement of the search result listing.

29. One or more computer-accessible media having computer-executable instructions embodied thereon that, when executed, cause a computing device to perform a method for displaying on a display device a search result Web page for a search term query, the search result Web page including at least two sections in which to display a search result listing, the at least two sections including a first section located in a more prominent area of the display device and the a second section located in a less prominent area of the display device, the method comprising:

assigning an expected performance of a search result listing based on a user defined expected click-through rate (CTR) that is adjusted based on a location of the section in the Web page in which it appears, wherein the expected performance of a listing is a threshold performance level that is tuned to a particular market;

said computing device measuring an actual performance of the search result listing, wherein the actual performance comprises at least one of CTR and context of the listing;

said computing device comparing the actual performance to the expected performance;

determining that the listing should be moved to the more prominent first section when the actual performance is better than the expected performance;

determining that the listing should be moved to the less prominent second section when the actual performance is poorer than the expected performance;

displaying the listing in the more prominent first section or in the less prominent second section, and

wherein assigning the expected performance includes increasing the expected performance when the listing is in the more prominent first section and decreasing the expected performance when the listing is in the less prominent second section, and

wherein determining that the listing should be moved to the more prominent or less prominent section is independent of an actual performance of any other listings.

30. The computer-accessible media of claim 29, wherein the more prominent first section includes the most relevant search result listings.

31. The computer-accessible media of claim 29, wherein the more prominent first section is located near a search term entry associated with the search result listings.

32. The computer-accessible media of claim 29, wherein the less prominent second section is located in a sidebar on the Web page.

33. The computer-accessible media of claim 29, wherein the less prominent second section includes less relevant search result listings than the more prominent first section.

34. The computer-accessible media of claim 29, further comprising increasing the expected performance of the listing based on a context of the listing, wherein the context of the listing comprises factors that increase performance, including at least one of a position of the listing above other listings, a larger size of the listing relative to other listings, a distinctive formatting applied to the display of the listing, including a contrasting color, a highlighting, an animation, and a graphic, and a good performance of a neighboring listing.

35. The computer-accessible media of claim 29, further comprising decreasing the expected performance of the listing based on a context of the listing, wherein the context of the listing comprises factors that decrease performance, including at least one of a position of the listing below other listings, a smaller size of the listing relative to other listings, a lack of distinctive formatting applied to the display of the listing, and a poor performance of a neighboring listing.

36. The computer-accessible media of claim 29, wherein the performance of a listing is a click-through rate, where the click-through rate is derived from a number of times the listing is displayed in the Web page as compared to a number of times the listing is clicked after being displayed.

37. The computer-accessible media of claim 29, wherein the listing is an unpaid listing, and the expected performance is a threshold level that is tuned to optimize unpaid listing relevance.

38. The computer-accessible media of claim 29, wherein the listing is paid listing that generates advertising revenue each time it is clicked, and the expected performance is a threshold level that is tuned to optimize paid listing revenue.

39. The computer-accessible media of claim 29, wherein the performance of a listing is based on an overall performance of the section in which the listing appears, and moving the listing includes moving all of the listings appearing in the section based on the overall performance.

40. The computer-accessible media of claim 39, wherein the overall performance of the first and second sections is based on the expected performance of each listing in the first and second sections, wherein the expected performance is based on where each listing appears within the first and second sections.

41. The computer-accessible media of claim 29, wherein the actual performance is better than the expected performance when the actual performance exceeds the expected performance, and poorer when the actual performance falls short of the expected performance.