Using natural search click events to optimize online advertising campaigns

• US 7,844,605 B2
• Filed: 04/20/2007
• Issued: 11/30/2010
• Est. Priority Date: 04/20/2007
• Status: Expired due to Fees

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Koran, Joshua M.
• Primary Examiner(s): Vital; Pierre M
• Assistant Examiner(s): Nofal; Christopher P
• Application Number:
  US11/738,195
• Publication Number:
  US 20080263025A1
• Time in Patent Office: 1,320 Days
• Field of Search: None
• US Class Current: 707/732
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0256 User search

Abstract

A method is described for augmenting sponsored search results in a search engine, which includes extracting attribute data from a plurality of natural searches for a search term linked to a plurality of uniform resource locators (URLs), analyzing the attribute data of one or more attributes for clickers and non-clickers to determine at least one greatest distinguishing factor between the clickers and non-clickers, and integrating the at least one greatest distinguishing factor into a matching algorithm used by the search engine to rank order and display a plurality of the most relevant ads corresponding to the plurality of URLs in response to a search for the term. The method may also integrate the at least one greatest distinguishing factor into a marketer algorithm to enable a marketer of a URL to strategically choose a search term, along with the at least one greatest distinguishing factor, on which to bid.

Claims

1. A computer-implemented method for augmenting sponsored search results in a search engine having a processor and storage, the method comprising:

extracting user attribute data by the search engine from a plurality of natural searches executed previously by a plurality of users for a search term that is linked to a plurality of uniform resource locators (URLs);

analyzing, by the search engine, the attribute data of one or more attributes for both clickers and non-clickers to determine at least one greatly distinguishing factor between the clickers and non-clickers for the one or more attributes;

grouping, by the search engine, the one or more attributes having at least a greatest distinguishing factor to identify one or more non-overlapping segments of clickers; and

in response to a search by a searching user for the term, integrating, by the search engine, the at least one greatest distinguishing factor into a matching algorithm to rank, order, and display a plurality of the most relevant ads corresponding to the plurality of URLs based on the aggregate of non-overlapping segments, wherein the plurality of most relevant ads have the highest aggregate of non-overlapping segment data matching the searching user.

2. The method of claim 1, further comprising:

integrating the at least one greatest distinguishing factor into a marketer algorithm used by the search engine to enable a marketer of a URL to strategically choose a search term, along with the at least one greatest distinguishing factor, on which to bid.

3. The method of claim 1, wherein the one or more attributes comprise user geography.

4. The method of claim 3, wherein the at least one greatest distinguishing factor comprises one or more selected from the group consisting of country, state, city, and zip code.

5. The method of claim 3, wherein the user geography is determined from internet Protocol (I.P.) addresses of users executing the natural searches.

6. The method of claim 1, wherein the one or more attributes comprise a user technographic, and the at least one greatest distinguishing factor comprises one or more selected form the group consisting of a connection speed, a type of browser, a size or quality of monitor, an operating system, and a plug-in of a connection device used by the user.

7. The method of claim 1, wherein the one or more attributes comprise a user demographic, and the at least one greatest distinguishing factor comprises one or more selected from the group consisting of age, gender, race, and disability.

8. The method of claim 1, wherein the one or more attributes comprise an aspect of user purchasing history, and the aspect of user purchasing history comprises one or more selected from the group consisting of a quantity purchased, a price paid per item, and a total price paid.

9. The method of claim 1, further comprising:

integrating the at least one greatest distinguishing factor into a marketer algorithm used by the search engine to suggest the search term to a marketer of a URL, along with the at least one greatest distinguishing factor, on which to bid.

10. A computer-implemented method for augmenting sponsored search results in a search engine having a processor and storage, the method comprising:

extracting user attribute data by the search engine from a plurality of natural searches executed previously by a plurality of users for a search term that is associated with a plurality of uniform resource locators (URLs);

analyzing, by the search engine, the data of a plurality of attributes for both clickers and non-clickers to determine at least one attribute that greatly differentiates clickers from non-clickers;

grouping, by the search engine, the clickers into different segments based at least on a greatest differentiating attribute;

identifying, by the search engine, one or more overlapping segments and non-overlapping segments;

calculating, by the search engine, relative click-through rates of each respective group based on the identified one or more overlapping and non-overlapping segments; and

presenting, by the search engine, a bidding interface that displays an optimum set of relative click-through rate results for each respective group based on the aggregate of one or more overlapping and non-overlapping segments.

11. The method of claim 10, further comprising:

integrating the at least one differentiating attribute into a matching algorithm used by the search engine to rank order and display at least one of the most relevant URL results and the most relevant ads corresponding to the plurality of URLs in response to a search for the term.

12. The method of claim 11, wherein the at least one differentiating attribute comprises one or more selected form the group consisting of a user geography, a user technographic, a user demographic, and a user purchasing history.

13. A computer-implemented method for augmenting sponsored search results in a search engine having a processor and storage, the method comprising:

extracting user attribute data, by the search engine, from a plurality of natural searches executed previously by a plurality of users for search terms that are linked to a plurality of uniform resource locators (URLs) and from web pages browsed by the plurality of users;

associating, by the search engine, the attribute data in memory with the specific users that conducted the natural searches and performed the web page browsing;

analyzing, by the search engine, the attribute data of one or more attributes for both clickers and non-clickers to determine at least one greatest distinguishing factor between the clickers and non-clickers for the one or more attributes;

grouping, by the search engine, the one or more attributes having the at least one greatest distinguishing factor to identify one or more overlapping segments of clickers;

integrating, by the search engine, the one or more overlapping segments of clickers into a matching algorithm used by the search engine to rank, order, and display a plurality of the most relevant ads corresponding to the plurality of URLs in response to a search for the search term;

executing, by the search engine, a targeting algorithm that incorporates the associated attribute data for each specific user that performed a search for a search term; and

re-ranking, by the search engine, a set of URL results based on the executed targeting algorithm and matching algorithm before displaying the results.

14. The method of claim 13, wherein the matching algorithm includes the targeting algorithm.

15. The method of claim 13, wherein the attribute data comprises one or more selected form the group consisting of a user geography, a user technographic, a user demographic, and a user purchasing history.

16. A computer-implemented system for augmenting sponsored search results in a search engine, the system comprising:

an index manager to extract user attribute data from a plurality of natural searches performed previously by a search engine on behalf of users for a search term that is linked to a plurality of uniform resource locators (URLs);

a database, in communication with the index manager, to store the extracted attribute data, wherein the database comprises user attributes associated in memory with specific users that performed the natural searches and aggregates of the user attributes grouped by type of attribute, wherein the database is stored in memory storage coupled with the search engine; and

a processor in communication with the database to;

analyze the user attributes for both clickers and non-clickers to determine at least one attribute that greatly differentiates clickers from non-clickers;

group the clickers into one or more overlapping segments based at least on a greatest differentiating attribute;

store the groups of clickers and overlapping segments in the database;

determine the relative click-through rates of the one of more groups of clickers based on the one or more overlapping segments; and

match a relevant ad or URL to a search user by aggregating the user attributes of the one or more overlapping segments.

17. The system of claim 16, wherein the memory storage includes a matching algorithm, wherein the processor integrates the at least one differentiating attribute into the matching algorithm used by the search engine to rank order and display the most relevant URL or ad results in response to a search for the term.

18. The system of claim 16, wherein the memory storage includes a marketer algorithm, wherein the processor integrates the at least one differentiating attribute into the marketer algorithm to enable a marketer of at least one URL to choose a strategic search term, along with the at least one differentiating attribute, on which to bid.

19. The system of claim 16, wherein the memory storage includes a targeting algorithm, wherein the processor:

executes the targeting algorithm that incorporates the associated attribute data for an identified user performing a search for a search term; and

re-ranks a set of URL results based on the executed targeting algorithm before being displayed to the identified user.

20. The system of claim 16, wherein the attribute types comprise user geography, and the at least one differentiating attribute comprises one or more selected form the group consisting of country, state, city, zip code, and Internet Protocol (I.P.) address.

21. The system of claim 16, wherein the attribute types comprise a user demographic, and wherein the at least one differentiating attribute comprises one or more selected form the group consisting of age, gender, race, and disability.

22. The system of claim 16, wherein the attribute types comprise an aspect of user purchasing history, which comprises one or more selected form the group consisting of a quantity of items purchased, a price paid per item, and a total price paid.

23. The system of claim 16, wherein the attribute types comprise a user technographic, and wherein the at least one differentiating attribute comprises one or more selected from the group consisting of a connection speed, a type of browser, a size or quality of monitor, an operating system, and a plug-in of a connection device used by the user.