Method and apparatus using historical influence for success attribution in network site activity

• US 8,793,236 B2
• Filed: 11/01/2012
• Issued: 07/29/2014
• Est. Priority Date: 11/01/2012
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Smith, Kevin G., George, William Brandon
• Primary Examiner(s): Coby, Frantz
• Application Number:
  US13/666,880
• Publication Number:
  US 20140122503A1
• Time in Patent Office: 635 Days
• Field of Search: 707/707, 707/821, 705/7.31, 705/14.1, 706/12, 706/45, 704/9
• US Class Current: 707/707
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/95 Retrieval from the web
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

User actions prior to, and associated with, an online success event may be considered participating actions that may have influenced the user toward the success event. A previously measured success influence metric for the participating actions may be used to determine a historical influence score for each participating action leading up to the success event. Each participating action may be assigned a current success influence score based on that event'"'"'s historical influence score as a percentage of a combined historical influence score of all the participating actions for the success event. Additionally, the assigned current success influence scores may be combined with the previously measured success influence metric for use as historical influence scores for success attribution regarding further instances of the success event.

Claims

1. A method, comprising:

performing, by a computer;

associating one or more participating actions with a success event;

wherein each of the one or more participating actions comprises a user action on a network-based content site;

determining a historical influence score for each of the one or more participating actions based on a previously measured metric associating the one or more participating actions with previous instances of the success event; and

assigning a current success influence score to each of the one or more participating actions based on the respective participating action'"'"'s historical influence score as a proportion of a total of the historical influence scores of the one or more participating actions.

2. The method of claim 1, wherein said assigning a current success influence score comprises attributing a portion of an overall success value for the success event to each of the one or more participating actions, and wherein the current success influence score indicates a level of contribution by the respective participating action toward the success event.

3. The method of claim 1, wherein the historical influence score of each respective participating action comprises an aggregate influence value for the respective participating action, and wherein the aggregate influence value is a combination of previous success influence values for the respective participating action.

4. The method of claim 3, further comprising combining the current success influence score of the respective participating action into the aggregate influence value for the respective participating action.

5. The method of claim 1, wherein the previously measured metric comprises, for each of the one or more participating actions, a corresponding conversion ratio;

wherein a respective conversion ratio comprises a ratio of a total number of times the corresponding participating action was previously associated with an instance of the success event as compared to a total number of previous instances of the success event.

6. The method of claim 1, wherein the previously measured metric comprises historical influence scores for participating actions associated with previous instances of the success event occurring only on the network-based content site.

7. The method of claim 1, wherein the previously measured metric comprises historical influence scores for participating actions associated with previous instances of the success event occurring on multiple network-based content sites.

8. The method of claim 1, wherein the historical influence score is based on previous instances of the success event that occurred within a predetermined age threshold.

9. A non-transitory, computer-readable storage medium storing program instructions that when executed on one or more computers cause the one or more computers to perform:

associating one or more user actions on a network-based content site with a success event;

determining a historical influence score for each of the one or more user actions based on a previously measured metric associating the one or more user actions with previous instances of the success event; and

assigning a current success influence score to each of one or more participating actions of the one or more user actions based on the respective participating action'"'"'s historical influence score as a proportion of a total of the historical influence scores of the one or more participating actions.

10. The medium of claim 9, wherein said assigning a current success influence score comprises attributing a portion of an overall success value for the success event to each of the one or more participating actions, and wherein the current success influence score indicates a level of contribution by the respective participating action toward the success event.

11. The medium of claim 9, wherein the success event comprises a web purchase of a product or service, and wherein the previously measured metric comprises historical influence scores associated with previous purchases of the product or service.

12. The medium of claim 9, wherein each of the one or more participating actions correspond to at least one of:

a web search, selecting a web link, visiting a web page, or entering data into a data entry field of the network-based content site.

13. The medium of claim 9, wherein the one or more user actions comprise a plurality of user actions, and wherein the program instructions further cause the one or more computers to perform:

determining that one or more of the plurality of user actions is a non-contributing action that does not contribute to the success event;

wherein said assigning a current success influence score is not performed for the one or more non-contributing actions.

14. The medium of claim 13, wherein said determining that one of the user actions is a non-contributing action comprises determining that the user action comprises a user checkout action that occurs prior to web purchases via the network-based content site, and wherein the success event is a web purchase via the network-based content site.

15. The medium of claim 13, wherein determining that one of the user actions is a non-contributing action comprises determining that the historical influence score for the user action is above a pre-determined occurrence threshold.

16. A system, comprising:

a data store storing historical influence data for a network-based site; and

one or more processors configured to implement a success influence analyzer, wherein the success influence analyzer is configured to;

associate one or more participating actions with a success event, wherein each participating action comprises a user action on a network-based content site;

access the historical influence data on the data store to determine a historical influence score for each of the one or more participating actions based on a previously measured metric associating the one or more participating actions with previous instances of the success event; and

assign a current success influence score to each of the one or more participating actions based on the respective participating action'"'"'s historical influence score as a proportion of a total of the historical influence scores of the one or more participating actions.

17. The system of claim 16, wherein to assign a current success influence score the success influence analyzer is configured to attribute a portion of an overall success value for the success event to each of the one or more participating actions, and wherein the current success influence score indicates a level of contribution by the respective participating action toward the success event.

18. The system of claim 16, wherein the success influence analyzer is further configured to update the data store according to the current success influence score of each of the one or more participating actions.

19. The system of claim 16, wherein the historical influence data on the data store comprises an aggregate influence value for each of the one or more participating actions and wherein the success influence analyzer is further configured to determine the historical influence score of each of the one or more participating actions based on the aggregate influence value for the respective participating action.

20. The system of claim 19, wherein the success influence analyzer is further configured to update the historical influence data on the data store to include an updated aggregate influence value for each of the one or more participating actions, wherein the updated aggregate influence value for each respective participating action is based on a combination of the current success influence score and the aggregate influence value for the respective participating action.