Method and Apparatus for Maximizing Publisher Revenue

• US 20110131099A1
• Filed: 12/01/2009
• Published: 06/02/2011
• Est. Priority Date: 12/01/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Yieldex, Inc. (Microsoft Corporation)
• Inventors: Cosman, C. Douglas, Shields, Tom, Miller, Craig, Barr, John
• Granted Patent: US 8,280,765 B2
• Time in Patent Office: Days
• US Class Current: 705/14.71
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0275 Auctions

Abstract

A computer implemented method for maximizing publisher revenue for online advertising is provided. The computer implemented method includes determining a ratio of guaranteed ad impressions to expected remaining ad impressions. If the ratio is less than a threshold, the method includes, identifying a minimum bid value, the minimum bid value based on the ratio; comparing the minimum bid value to a received bid value for an ad impression; allocating the ad impression to an owner of the received bid when the received bid value is greater than or equal to the minimum bid value; and displaying an ad of the owner of the received bid. If the ratio is greater than the threshold, the method includes, allocating the ad impression to an owner of the guaranteed ad impressions. In one embodiment, the method operations are stored on a computer readable storage medium.

Claims

1. A computer implemented method for maximizing publisher revenue for targeted advertising, comprising:

determining a ratio of guaranteed ad impressions to expected remaining ad impressions;

if the ratio is less than a threshold value, the method includes,

identifying a minimum bid value, the minimum bid value based on the ratio;

comparing the minimum bid value to a received bid value for an ad impression;

allocating the ad impression to an owner of the received bid when the received bid value is greater than or equal to the minimum bid value; and

displaying an ad of the owner of the received bid.

2. The method of claim 1, further comprising:

allocating the ad impression to the owner of the guaranteed ad impressions when one of the received bid value is less than the minimum bid value or if the ratio is greater than the threshold.

3. The method of claim 1, wherein each method operation is performed in real time through an ad server and wherein the received bid value is a highest bid value selected from a plurality of received bid values.

4. The method of claim 1, wherein identifying a minimum bid value includes, applying the ratio to a normal distribution of bid values for all ad impressions.

5. The method of claim 4, wherein all the ad impressions include the guaranteed ad impressions and non-guaranteed ad impressions.

6. The method of claim 1, further comprising:

recalculating the ratio for each request for an ad to be displayed.

7. The method of claim 1, wherein identifying a minimum bid value includes,

maintaining a distribution sample of bid values of all ad impressions;

identifying a bid value of the distribution sample correlated to the ratio, the bid value determining the minimum bid value; and

recalculating the ratio after allocating the ad impression to one of the owner of the received bid or the owner of the guaranteed ad impressions.

8. The method of claim 7, further comprising:

updating the distribution sample after allocating the ad impression; and

repeating the identifying the bid value and the recalculating the ratio for the updated distribution type.

9. The method of claim 8, further comprising:

maintaining a history of bid values and corresponding allocations resulting from the history of bid values, wherein the updated distribution sample is defined through the history of bid values and wherein the updated distribution sample is different from the distribution sample of bid values.

10. A computer implemented method for guaranteeing a number of impressions utilizing a non-guaranteed mechanism for targeted advertising, comprising:

receiving a request for supplying a bid for an ad impression;

submitting a bid value in response to receiving the request, the submitting including,

identifying a minimum bid value based on a ratio of guaranteed ad impressions to expected remaining ad impressions;

setting the minimum bid value to the bid value;

identifying a highest bid value from all received bid values; and

allocating the ad impression to an owner of the highest bid value; and

displaying an ad of the owner of the received bid.

11. The method of claim 10, wherein each method operation is performed in real time for displaying the ad within a browser of a computing device.

12. The method of claim 10, wherein identifying a minimum bid value includes, applying the ratio to a normal distribution of bid values for all available ad impressions within a segment.

13. The method of claim 10, wherein the identifying includes,

maintaining a distribution type of bid values of all ad impressions within a particular segment;

identifying a bid value of the distribution type correlated to the ratio, the bid value determining the minimum bid value; and

recalculating the ratio after allocating the ad impression.

14. The method of claim 13, further comprising:

updating the distribution type after allocating the ad impression.

15. The method of claim 10, wherein the identifying the highest bid and the allocating the ad impression are performed through an ad exchange, while the receiving and the submitting are performed by an ad bidder in communication with the ad exchange and wherein the ad bidder is one of a plurality of ad bidders in communication with the ad exchange.

16. The method of claim 14, wherein the updating the distribution type includes,

capturing a history of bid values for the particular segment;

and

applying a curve fitting model to the history of bid values, wherein output of the curve fitting model is associated with a different distribution type.

17. A computer implemented method for maximizing publisher revenue for targeted advertising, comprising:

setting an initial minimum value;

allocating ad impressions to bids exceeding the initial minimum value;

displaying the ad impressions for the bids exceeding the initial minimum value;

tracking the bid values of ad impressions over a time period;

adjusting the initial minimum value according to the tracking;

allocating subsequent ad impressions to bids exceeding the adjusted minimum value; and

displaying the ad impressions for the bids exceeding the adjusted initial minimum value.

18. The method of claim 17, wherein the tracking the bid values includes, trending an average cost per unit of ad impressions over the time period for the initial minimum value, and wherein the adjusting the initial minimum value includes, adjusting the initial minimum value according to a direction determined by the trending.

19. The method of claim 18, further comprising:

increasing the initial minimum value when the trending is increasing; and

decreasing the initial minimum value is decreasing.

20. The method of claim 19, wherein one of the increasing or decreasing is performed by an amount proportional to a change in the trending for the time period.

21. The method of claim 17, wherein each method operation is embodied as computer instructions on a computer readable medium.

22. A computer implemented method for maximizing publisher revenue for targeted advertising, comprising:

setting an initial minimum value;

allocating ad impressions to bids exceeding the initial minimum value;

displaying the ad impressions for the bids exceeding the initial minimum value;

tracking the bid values of ad impressions over a time period;

adjusting the initial minimum value according to one of the tracking or external data;

allocating subsequent ad impressions to bids exceeding the adjusted minimum value; and

displaying the ad impressions for the bids exceeding the adjusted initial minimum value.