Simulating advertising campaigns

• US 10,565,613 B2
• Filed: 03/27/2015
• Issued: 02/18/2020
• Est. Priority Date: 03/27/2015
• Status: Expired due to Fees

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Sodomka, Eric Michael, Bailey, Michael Charles
• Primary Examiner(s): Detweiler, James M
• Assistant Examiner(s): Mandel, Monica A
• Application Number:
  US14/671,991
• Publication Number:
  US 20160283974A1
• Time in Patent Office: 1,789 Days
• CPC Class Codes: G06Q 30/0247 Calculate past, present or future revenues
• CPC Class Codes: G06Q 30/0275 Auctions

Abstract

An advertising simulation system performs simulation of advertising results. The advertising results indicate the results of a simulated auction for a set of ad queries that are provided advertisements from available advertisements. The simulated auction applies an advertising model to perform an auction using advertising conditions. As advertisers are selected to provide advertisements to individual ad queries, advertising conditions, such as advertiser'"'"'s bidding amounts, may change. To effectively simulate a large number of ad queries, the advertising segments simulation and performs simulation of each segment. Each segment'"'"'s ad results may affect the conditions of a subsequent segment, and the change in conditions may change the size of a subsequent segment.

Claims

1. A method comprising:

identifying a set of previously received, time-ordered advertising queries, each advertising query in the set of previously received advertising queries identifying a user and defining a past opportunity to present at least one advertisement to the user;

determining, by a processor, simulation auction results for the set of advertising queries as a plurality of sequential segments of advertising queries in the set of advertising queries, by;

determining advertising results for a first segment of the plurality of sequential segments by assigning advertising queries to an advertiser from a set of advertisers and by applying advertising conditions to an advertising model for the first segment of advertising queries, the advertising results including an auction winner and winner price;

sequentially determining advertising results for each additional segment of the plurality of sequential segments by;

updating one or more advertising conditions for each advertiser from the set of advertisers based on the determined results of prior segments, the advertising conditions including a paced bid amount;

determining a change in advertising conditions for at least one advertiser in the set of advertisers by comparing the updated advertising conditions of the additional segment to the advertising conditions of a prior segment;

determining a segment size of advertising queries for the additional segment based on the change in advertising conditions compared to a threshold;

determining advertising results for the additional segment having the determined segment size by assigning each advertising query in the additional segment to an advertiser from the set of advertisers by applying the updated advertising conditions to the advertising model for the additional segment of advertising queries, wherein the advertising conditions are constant while determining the advertising results for the additional segment.

2. The method of claim 1, wherein determining the segment size of advertising queries for the additional segment comprises determining the change in the paced bid amount for the at least one advertiser in the set of advertisers, and increasing the segment size of advertising queries for the additional segment when the change in paced bid amount is below a threshold value or decreasing the segment size of advertising queries for the additional segment when the change in paced bid amount is above a threshold value.

3. The method of claim 2, wherein determining the change in paced bid amount comprises determining the change in paced bid amount from a plurality of advertising results associated with a plurality of sequential segments of advertising queries.

4. The method of claim 1, wherein updating advertising conditions comprise updating the paced bid amount for at least one advertiser from the set of advertisers based on a budget for the advertiser.

5. The method of claim 1, further comprising:

determining alternate advertising results for another segment of the plurality of sequential segments by applying advertising conditions to an alternate advertising model for the first segment of the plurality of sequential segments.

6. The method of claim 1, wherein each of the advertising queries is associated with a reserve price, and further comprising:

identifying a modified reserve price associated with at least one advertising query; and

determining alternate advertising results for another segment of the plurality of sequential segments by applying the modified reserve price to the advertising model for the first segment of the plurality of sequential segments.

7. The method of claim 1, wherein the paced bid amounts for the advertisers are determined based on a pacing algorithm, and further comprising:

applying a modified pacing algorithm to determine modified paced bid amount for at least one advertiser;

determining alternate advertising results for another segment of the plurality of sequential segments by applying the modified paced bid amount to the advertising model for the first segment of the plurality of sequential segments.

8. A non-transitory computer-readable medium storing instructions for execution on a processor, the instructions when executed by the processor causing the processor to perform steps comprising:

identifying a set of previously received, time-ordered advertising queries, each advertising query in the set of previously received advertising queries identifying a user and defining a past opportunity to present at least one advertisement to the user;

determining, by a processor, simulation auction results for the set of advertising queries as a plurality of sequential segments of advertising queries in the set of advertising queries, by;

determining advertising results for a first segment of the plurality of sequential segments by assigning advertising queries to an advertiser from a set of advertisers and by applying advertising conditions to an advertising model for the first segment of advertising queries, the advertising results including an auction winner and winner price;

sequentially determining advertising results for each additional segment of the plurality of sequential segments by;

updating one or more advertising conditions for each advertiser from the set of advertisers based on the determined results of prior segments, the advertising conditions including a paced bid amount;

determining a change in advertising conditions for at least one advertiser in the set of advertisers by comparing the updated advertising conditions of the additional segment to the advertising conditions of a prior segment;

determining a segment size of advertising queries for the additional segment based on the change in advertising conditions compared to a threshold;

determining advertising results for the additional segment having the determined segment size by assigning each advertising query in the additional segment to an advertiser from the set of advertisers by applying the updated advertising conditions to the advertising model for the additional segment of advertising queries, wherein the advertising conditions are constant while determining the advertising results for the additional segment.