Optimization of Content Placement

• US 20130060629A1
• Filed: 09/07/2011
• Published: 03/07/2013
• Est. Priority Date: 09/07/2011
• Status: Abandoned Application

• Current Assignee: Truex Media Inc.
• Original Assignee: Truex Media Inc.
• Inventors: Rangsikitpho, Joshua, Stein, Aaron
• Application Number:
  US13/226,654
• Publication Number:
  US 20130060629A1
• Time in Patent Office: Days
• US Class Current: 705/14.42
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Systems and methods for optimizing the placement of online advertisements based on information describing user interactions with previous advertisement. The information include primary activity data describing user responses that directly caused the previous advertisements to pay and secondary activity data describing user responses to the previous advertisements that did not directly cause the previous advertisements to pay. A performance ratings of candidate advertisements are determined based upon the primary activity data and secondary activity data. An advertisement from the candidate advertisements for the placement opportunity is selected based on their performance ratings

Claims

1. A computer-implemented system for optimizing the placement of online advertisements, the system comprising at least one processor programmed to execute an optimization engine, wherein the optimization engine is programmed to:

receive placement data indicating a placement opportunity to serve an advertisement to a client device of a user;

access available advertisement data indicating a plurality of available advertisements, wherein the available advertisements comprise advertisements that pay according to a plurality of different payment types;

access summarized, primary activity data of user responses to a plurality of previous advertisements, wherein the primary activity data is derived from user responses to the previous advertisement that directly caused the previous advertisements to pay;

access summarized, secondary activity data of user responses to the plurality of previous advertisements, wherein the secondary data is derived from user activities associated with the previous advertisements and occurring after the presentation of the previous advertisement, but that did not directly cause the previous advertisements to pay;

generate a performance rating for each of the plurality of candidate advertisements based upon the summarized primary activity data and the summarized secondary activity data; and

select an advertisement for the placement opportunity from the plurality of candidate advertisements based on the performance ratings for each of the plurality of candidate advertisements.

2. The system of claim 1, wherein the plurality of payment types comprises cost-per-engagement (CPE), cost-per-action (CPA), and cost-per impression (CPM).

3. The system of claim 1, wherein the at least one processor is also programmed to execute a tracking engine, wherein the tracking engine is programmed to:

access the primary activity data;

categorize the primary activity data into summarized primary activity data;

access secondary activity data;

categorize the secondary activity data into summarized secondary activity; and

store the summarized primary and secondary activity data to a data store.

4. The system of claim 3, wherein to categorize the primary activity data comprises to categorize based on at least one factor selected from the group consisting of:

demographic data describing the users, payment types for the previous advertisements, goods promoted by the previous advertisements, services promoted by the previous advertisements, content distributors through which the previous advertisements were served, and partner networks through which at least a portion of the previous advertisements were served.

5. The system of claim 1, wherein to generate the performance rating for each of the plurality of candidate advertisements, the optimization engine is further programmed to determine an expected revenue for the candidate advertisement.

6. The system of claim 5, wherein, to determine the expected revenue for the candidate advertisement, the optimization engine is further programmed to:

conditioned upon the candidate advertisement being a cost-per-impression advertisement, determine the expected revenue based upon a reimbursement rate per impression for the candidate advertisement.

7. The system of claim 5, wherein, to determine the expected revenue for the candidate advertisement, the optimization engine is further programmed to:

conditioned upon the candidate advertisement paying based on a primary user activity, determine a likelihood that the advertisement will pay based the summarized primary activity data for other users similar to the user for other advertisements similar to candidate advertisement paying based on the same primary user activity; and

determine the expected revenue for the candidate advertisement based upon the determined likelihood.

8. The system of claim 7, wherein, to determine the expected revenue for the candidate advertisement further, the optimization engine is further programmed to:

determine a further likelihood that the advertisement will pay based the summarized secondary activity data for other users similar to the user for other advertisements similar to candidate advertisement; and

determine the expected revenue for the candidate advertisement based upon the further likelihood.

9. The system of claim 7, wherein other users similar to the user comprise at least one of users with similar demographic backgrounds, users who were served the previous advertisements via a common content distributor, and users who were served the previous advertisements via a common partner network.

10. The system of claim 1, wherein the optimization engine is further programmed to remove from the plurality of candidate advertisements any advertisements that cannot be served to the user due to a cap constraint.

11. The system of claim 10, wherein the cap constraint comprises at least one limitation selected by the optimization engine from the group consisting of:

a limitation on a number of times that an advertisement can be shown to a single user;

a limitation on a number of times that an advertisement can be shown to users served the advertisement by a common content distributor;

a limitation that specifies a required mix of advertisement payment types for advertisements served to the user; and

a limitation that specifies a required mix of advertisement formats for advertisements served to the user.

12. The system of claim 11, wherein the required mix of advertisement payment types for advertisements served to the user comprises a distribution of cost-per-action (CPA) advertisements, cost-per-engagement (CPE) advertisements, and of cost-per-impression advertisements.

13. The system of claim 1, wherein the optimization engine is further programmed to:

determine whether a schedule constraint applies to any of the plurality of candidate advertisements, wherein the schedule constraint defines a predetermined number of paying placements of the advertisement to occur within a predetermined time; and

conditioned on a schedule constraint applying to a first advertisement selected from the candidate advertisements, apply a weight to the performance rating for the first advertisement.

15. A computer-implemented method for optimizing the placement of online advertisements, the method comprising:

receiving placement data indicating a placement opportunity to serve an advertisement to a client device of a user;

accessing available advertisement data indicating a plurality of available advertisements, wherein the available advertisements comprise advertisements that pay according to a plurality of different payment types;

accessing summarized, primary activity data of user responses to a plurality of previous advertisements, wherein the primary activity data is derived from user responses to the previous advertisement that directly caused the previous advertisements to pay;

accessing summarized, secondary activity data of user responses to the plurality of previous advertisements, wherein the secondary data is derived from user activities associated with the previous advertisements and occurring after the presentation of the previous advertisement, but that did not directly cause the previous advertisements to pay;

generating a performance rating for each of the plurality of candidate advertisements based upon the summarized primary activity data and the summarized secondary activity data; and

selecting an advertisement for the placement opportunity from the plurality of candidate advertisements based on the performance ratings for each of the plurality of candidate advertisements.

14. The system of claim 15, wherein to apply the weight applied to the performance rating for the first advertisement, the optimization engine is further programmed to select the weight based on a number of previous paying placements of the advertisement resulting in payment and an amount of time remaining prior to the end of the predetermined time.