Adjusting Bids Based on Predicted Performance

• US 20100138291A1
• Filed: 11/30/2009
• Published: 06/03/2010
• Est. Priority Date: 12/02/2008
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Corduneanu, Adrian Dumitru, Gallagher, James A., Benson, Scott S., Silverman, Andrew E., Llinares, Tomas Lloret, Sharma, Abhinay, Pfleger, Karl Robert
• Granted Patent: US 8,271,325 B2
• Time in Patent Office: Days
• US Class Current: 705/14.450
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0242 Determining effectiveness of advertisements
• CPC Class Codes: G06Q 30/0246 Traffic
• CPC Class Codes: G06Q 30/0247 Calculate past, present or future revenues
• CPC Class Codes: G06Q 30/0275 Auctions

Abstract

A predicted performance metric of a candidate advertisement in an advertising slot can be received and compared to a baseline predicted performance metric for the candidate advertisement. A target bid associated with the candidate advertisement can be adjusted based upon the comparison, and the adjusted bid can be submitted to an auction for the advertising slot.

Claims

1. A computer implemented method, comprising:

receiving a predicted performance metric for a candidate advertisement in an advertising slot;

comparing the predicted performance metric for the candidate advertisement in the advertising slot to a baseline predicted performance metric of the candidate advertisement;

adjusting a target bid associated with the candidate advertisement based upon the comparison of the predicted performance metric for the candidate advertisement and the baseline predicted performance metric for the candidate advertisement; and

submitting the adjusted target bid to an auction for the advertising slot.

2. The computer implemented method of claim 1, wherein the predicted performance metric is based upon historical conversion tracking data for previous advertising slots.

3. The computer implemented method of claim 2, wherein the predicted performance metric is further based upon historical conversion tracking data for the current advertisement in past advertising slots similar to the current advertising slot.

4. The computer implemented method of claim 1, further comprising identifying the baseline predicted performance metric.

5. The computer implemented method of claim 4, wherein identifying the baseline predicted performance metric comprises:

receiving predicted conversion rates for a set of previous auctions associated with the candidate advertisement; and

determining an average predicted conversion rate from the set of previous auctions and associated predicted conversion rates;

wherein the baseline predicted performance metric comprises the average predicted conversion rate.

6. The computer implemented method of claim 5, further comprising identifying those previous auctions in which the candidate advertisement received a click as the set of previous auctions for which predicted conversion rates are received.

7. The computer implemented method of claim 5, further comprising identifying those previous auctions in which the candidate advertisement received a conversion as the set of previous auctions for which predicted conversion rates are received.

8. The computer implemented method of claim 5, further comprising identifying those previous auctions in which the candidate advertisement was served as the set of previous auctions for which predicted conversion rates are received.

9. The computer implemented method of claim 5, further comprising identifying those previous auctions in which the candidate advertisement was also a candidate advertisement as the set of previous auctions for which predicted conversion rates are received.

10. The computer implemented method of claim 1, wherein:

comparing the predicted performance metric of the candidate advertisement to the baseline predicted performance metric for the candidate advertisement comprises dividing the predicted performance metric by the baseline predicted performance metric to derive an adjustment ratio; and

adjusting the target bid comprises multiplying the target bid by the adjustment ratio.

11. The computer implemented method of claim 1, further comprising:

determining an average price paid for placement of the advertisement;

comparing the average price paid to the target bid;

if the average price paid exceeds the target bid, throttling the adjusted bid.

12. The computer implemented method of claim 11, wherein throttling the adjusted bid comprises decreasing the adjusted bid to the target bid.

13. The computer implemented method of claim 11, wherein throttling the adjusted bid comprises decreasing the adjusted bid to a maximum adjusted bid.

14. The computer implemented method of claim 11, wherein throttling the adjusted bid comprises discounting the adjusted bid by a percentage based on how much the average price paid exceeds the target bid.

15. A computer-readable medium having instructions stored thereon, which, when executed by a processor, causes the processor to perform operations comprising:

receiving a predicted conversion rate for a candidate advertisement in the current advertising slot;

comparing the predicted conversion rate for the advertisement in the current advertising slot to a baseline predicted conversion rate;

adjusting the target bid associated with the advertisement based upon the comparison of the predicted conversion rate and the baseline predicted conversion rate; and

submitting the adjusted target bid to the impending auction.

16. The computer-readable medium of claim 15, wherein the predicted conversion rate is based upon historical conversion tracking data for the current advertisement in past advertising slots similar to the current advertising slot.

17. The computer-readable medium of claim 15, further comprising identifying the baseline predicted conversion rate.

18. The computer-readable medium of claim 17, wherein identifying the baseline predicted conversion rate comprises:

identifying a set of previous auctions;

receiving predicted conversion rates for the set of previous auctions; and

determining an average predicted conversion rate from the set of previous auctions and associated predicted conversion rates;

wherein the baseline predicted conversion rate comprises the average predicted conversion rate.

19. The computer-readable medium of claim 15, wherein comparing the predicted conversion rate to the baseline rate comprises dividing the predicted conversion rate by the baseline predicted conversion rate to derive an adjustment ratio, and wherein adjusting the target bid comprises multiplying the target bid by the adjustment ratio.

20. The computer-readable medium of claim 15, further comprises:

determining an average discount bid associated with the advertisement, wherein the average discount bid is the average amount paid by the advertiser for the advertisement;

comparing the average discount bid to the target bid; and

if the average discount bid exceeds the target bid, throttling the adjusted bid.

21. A computer implemented method, comprising:

receiving a predicted click-through rate for a candidate advertisement in an available advertising slot;

comparing the predicted click-through rate for the candidate advertisement in the available advertising slot to an average predicted click-through rate for the candidate advertisement derived from a control set; and

adjusting a target impression based bid associated with the candidate advertisement based upon the comparison of the predicted click-through rate for the candidate advertisement and the average predicted click-through rate for the candidate advertisement;

submitting the adjusted target impression based bid to an auction for the available advertising slot.

22. A computer implemented method, comprising:

receiving a predicted revenue metric for a candidate advertisement in an available advertising slot;

comparing the predicted revenue metric for the advertisement in the available advertising slot to an average predicted revenue metric for the candidate advertisement derived from a control set; and

adjusting a target conversion based bid associated with the candidate advertisement based upon the comparison of the predicted revenue metric for the candidate advertisement and the average predicted revenue metric for the candidate advertisement;

submitting the adjusted target bid to an auction for the available advertising slot.