Automatic bid adjustments for electronic advertising

• US 8,719,082 B1
• Filed: 11/10/2008
• Issued: 05/06/2014
• Est. Priority Date: 11/10/2008
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Brandt, David Richard, Herrmann, Eric Alfred, Snyder, Luke A.
• Primary Examiner(s): Lastra, Daniel
• Application Number:
  US12/267,870
• Time in Patent Office: 2,003 Days
• Field of Search: 705/14.1
• US Class Current: 705/14.1
• CPC Class Codes: G06Q 30/0275 Auctions
• CPC Class Codes: G06Q 90/00 Systems or methods specially adapted for administrative, commercial, financial, managerial or supervisory purposes, not involving significant data processing

Abstract

The accuracy of bid amounts for electronic advertising is improved by accounting for factors such as the cost-per-click (CPC) ratio for each ad. When a provider such as a search engine selects advertisements using automated auctions, it can be desirable for an advertiser to avoid underbidding for ads when the ads generate a significant amount of revenue or profit, or are otherwise performing well. Various algorithms can be used to generate a bid adjustment factor that allows bid values to be increased (or decreased) as appropriate, based on information such as CPC information. By calculating a separate adjustment factor, the amount of adjustment can be monitored and/or capped to avoid overspending. The algorithms also can utilize information at various levels of categorization, and at different time intervals, depending on the amount and type of information available, in order to provide an accurate and significant result.

Claims

1. A method in a computer system for generating a bid price for an electronic advertisement, comprising:

under the control of one or more computer systems configured with executable instructions,

selecting a keyword and a provider with which to place an advertisement, the provider operable to generate a page including content associated with the keyword;

determining a first set of bidding data components related to the advertisement, each bidding data component containing data stored at a different level in a bidding data hierarchy related to an aspect of the advertisement;

determining a minimum number of data points needed to determine a significant bidding result;

selecting a second set of bidding components including a bidding data component at a lowest level in the first set of bidding components and including a number of data points at least meeting the minimum number of data points, the second set of bidding components including at least one bidding component at a next highest level in the bidding data hierarchy when the bidding data component at the lowest level does not include the minimum number of data points, the bidding data component at the lowest level having a first targeted data definition, the at least one bidding data component at the next highest level having a second targeted data definition of which the first targeted data definition is a subset, the second targeted data definition being broader than the first targeted data definition;

selecting at least one time period over which to analyze data for each bidding component;

determining a base bid price by analyzing first bid data for each of the second set of bidding components for each selected time period;

determining a bid adjustment factor by analyzing second bid data for each of the second set of bidding components for each selected time period, the second bid data including previously submitted bid prices and actual costs incurred as a result of the previously submitted bid prices;

determining a bid result by adjusting the base bid price by the bid adjustment factor; and

submitting advertising information and the bid result to the provider for the keyword.

2. A method according to claim 1, further comprising:

weighting each bidding data component based at least in part on a number of data points for the data component compared to a number of data points remaining to reach the minimum number of data points.

3. A method according to claim 1, further comprising:

weighting each time period based at least in part upon a recency of the time period.

4. A method in a computer system for adjusting a bid price in an electronic environment, comprising:

under the control of one or more computer systems configured with executable instructions,

selecting a set of bidding components from a bidding data hierarchy, the set of bidding components including a bidding data component at a lowest level in the bidding data hierarchy and a number of bidding data components at next highest levels in the bidding data hierarchy needed to at least meet a minimum number of data points, the bidding data component at a lowest level having a data definition, the bidding data components at a next highest level in the hierarchy having a broader data definition of which lower level data definitions in the hierarchy are a subset;

selecting at least one time period over which to analyze data for each bidding component in the set;

determining a bid adjustment factor by analyzing bid data for each of the set of bidding components for each selected time period, the bid data including previously submitted bid prices and actual costs incurred as a result of the previously submitted bid prices;

determining a bid result by adjusting a base bid price by the bid adjustment factor; and

submitting the bid result to a provider operable to receive bids electronically.

5. A method according to claim 4, further comprising:

generating the base bid price.

6. A method according to claim 5, wherein:

the base bid price is generated by analyzing bid data for each of the set of bidding components for each selected time period.

7. A method according to claim 4, wherein:

each bidding data component is weighted based at least in part on a number of data points for that data component compared to a number of data points remaining to reach the minimum number of data points.

8. A method according to claim 4, wherein:

each time period is weighted based at least in part upon a recency of the time period.

9. A method according to claim 8, further comprising:

normalizing the weighted time periods before determining the bid adjustment factor.

10. A method according to claim 4, wherein:

the bid price is generated for an electronic advertisement to be generated for display by the provider.

11. A method according to claim 10, wherein:

the bid data analyzed for the bid adjustment factor includes a cost-per-click ratio for the advertisement and the provider.

12. A method according to claim 4, wherein:

each time period is a decay value having a half life and wherein analyzing bid data for the time period includes weighting corresponding data based at least in part on the half life.

13. A method according to claim 4, further comprising:

determining whether the bid adjustment factor exceeds a specified threshold for the adjustment; and

if the bid adjustment factor exceeds the specified threshold, setting the specified threshold as the bid adjustment factor.

14. A method according to claim 4, further comprising:

determining the minimum number of data points for the bid result to be of significance.

15. A method according to claim 4, further comprising:

preventing the base bid price from being adjusted if there is no data for any of the set of bidding components within a specified period of recency.

16. A method according to claim 4, further comprising:

storing data to the data hierarchy received as a result of submitting the bid result to the provider.

17. A system for adjusting a bid price in an electronic environment, comprising:

a processor; and

a memory device including instructions that, when executed by the processor, cause the processor to;

select bidding data components from a hierarchy of bidding data, including a bidding data component at a first level in the hierarchy and a number of bidding data components at next highest levels in the hierarchy needed to at least meet a specified number of data points for significance, the bidding data component at the first level having a data definition, the bidding components at next highest levels having subsequently broader data definitions of which lower level data definitions are a subset of a next highest level data definition in the hierarchy;

select at least one time period over which to analyze data for each selected bidding data component;

determine a bid adjustment factor by analyzing bid data for each of the bidding data components for each selected time period, the bid data including previously submitted bid prices and corresponding actual costs;

determine a bid result by multiplying the base bid price by the bid adjustment factor; and

submit the bid result to a provider operable to receive bids electronically.

18. A system according to claim 17, wherein the instructions, when executed by the processor, further cause the processor to:

generate the base bid price by analyzing bid data for each of the bidding components for each selected time period.

19. A system according to claim 17, wherein the instructions, when executed by the processor, further cause the processor to:

weight each bidding data component based at least in part on a number of data points for that bidding data component compared to a number of data points remaining to reach the minimum number of data points.

20. A system according to claim 17, wherein the instructions, when executed by the processor, further cause the processor to:

weight each time period based at least in part upon a recency of the time period.

21. A computer program product embedded in a computer readable storage medium for adjusting a bid price for an advertisement in an electronic environment, comprising:

program code executable by one or more processors for selecting bidding data components from a hierarchy of bidding data, including a bidding data component at a first level in the hierarchy and a number of bidding data components at next highest levels in the hierarchy needed to at least meet a specified number of data points for significance, the bidding data component at a first level having a data definition, the bidding components at a next highest level having broader data definitions than lower level data definitions, the lower level data definitions forming a subset of the next highest level in the hierarchy;

program code executable by the one or more processors for selecting at least one time period over which to analyze data for each selected bidding data component;

program code executable by the one or more processors for determining a bid adjustment factor by analyzing bid data for each of the bidding data components for each selected time period, the bid data including previously submitted bid prices and corresponding actual costs;

program code executable by the one or more processors for determining a bid result by multiplying the base bid price by the bid adjustment factor; and

program code executable by the one or more processors for submitting the bid result to a provider operable to receive bids electronically.

22. A computer program product according to claim 21, further comprising:

program code executable by the one or more processors for generating the base bid price by analyzing bid data for each of the bidding components for each selected time period.

23. A computer program product according to claim 21, further comprising:

program code executable by the one or more processors for weighting each bidding data component based at least in part on a number of data points for that bidding data component compared to a number of data points remaining to reach the minimum number of data points.

24. A computer program product according to claim 21, further comprising:

program code executable by the one or more processors for weighting each time period based at least in part upon a recency of the time period.

25. A computer program product according to claim 21, wherein:

the bid data analyzed for the bid adjustment factor includes a cost-per-click ratio for the advertisement.