Systems and methods for determining competitive market values of an ad impression

• US 10,223,703 B2
• Filed: 07/18/2011
• Issued: 03/05/2019
• Est. Priority Date: 07/19/2010
• Status: Active Grant

• Current Assignee: Mediamath Acquisition Corporation
• Original Assignee: Mediamath LLC
• Inventors: Buchalter, Yehuda Ari
• Primary Examiner(s): Li, Sun M
• Assistant Examiner(s): Nguyen, Thuy N
• Application Number:
  US13/185,236
• Publication Number:
  US 20120041816A1
• Time in Patent Office: 2,787 Days
• Field of Search: 705 14
• 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 30/0275 Auctions

Abstract

The present disclosure is directed to methods and systems for determining competitive market values for an ad impression on an advertiser exchange. An engine executing on a device may receive a candidate set of inputs associated with ad impressions. The engine may determine competitive market values for an ad impression on an advertiser exchange. The engine may determine candidate clearing prices based on the candidate set of inputs and history of clearing prices on the advertiser exchange. The engine may generate, based on the candidate clearing prices, a competitive market value prediction for the ad impression on the advertiser exchange. The competitive market value prediction may comprise a distribution function of predicted clearing prices on the advertiser exchange. The engine may generate, based on the competitive market value prediction, a fair market value bid for the ad impression in the context of a specific ad campaign.

Claims

1. A computer-implemented method for determining competitive market value predictions for a plurality of ad impressions across a plurality of ad impression exchanges, the method comprising:

receiving, by an engine executing on a computer system, a candidate set of inputs from an advertiser defining two or more criteria for selecting the plurality of ad impressions, the engine configured to generate a competitive market value prediction for the plurality of ad impressions on the plurality of ad impression exchanges;

receiving, by the engine, an aggressiveness value representative of a level of aggressiveness of the advertiser for bidding on the plurality of ad impressions;

receiving, by the engine, one or more targets desired by the advertiser for the plurality of ad impressions, wherein the one or more targets comprise one or more of a targeting segment, channel, or daypart;

dynamically selecting, by the engine, a subset of the candidate set of inputs for determining candidate clearing prices for the plurality of ad impressions on the plurality of ad impression exchanges, wherein the dynamically selecting the subset of the candidate set of inputs is based at least in part on a profile of one or more past successful bids, wherein the candidate set of inputs comprises two or more selected from a group consisting of;

a campaign goal of the advertiser, a budget of an ad campaign of the advertiser, performance of an ad campaign of the advertiser, or a spending pace for the budget;

receiving, by the engine, data relating to historical clearing prices for the plurality of ad impressions on the plurality of ad impression exchanges, each of the historical clearing prices comprising an ad impression bid price that has been accepted on the plurality of ad impression exchanges;

determining, by the engine, the candidate clearing prices, wherein the determining the candidate clearing prices further comprises;

performing, by a single interface of the computer system, normalization of data transmission interfaces of each of the plurality of ad impression exchanges, wherein the normalization further comprises;

accessing the data transmission interfaces of each of the plurality of ad impression exchanges; and

translating data relating to the candidate set of inputs and the historical clearing prices among the data transmission interfaces of each of the plurality of ad impression exchanges into the single interface; and

selecting, by the engine, a subset of the historical clearing prices, wherein the selecting is performed by using the selected subset of the candidate set of inputs, wherein the selected subset of the historical clearing prices correspond to the candidate clearing prices, wherein each of the candidate clearing prices is for an ad impression corresponding to the selected subset of the candidate set of inputs;

generating, by the engine based at least in part on the candidate clearing prices, the competitive market value prediction for the plurality of ad impressions on the plurality of ad impression exchanges, the competitive market value prediction comprising a distribution function of bid prices and probabilities of each of the bid prices being accepted on the plurality of ad impression exchanges for the plurality of ad impressions;

dynamically adjusting, by the engine, the generated competitive market value prediction for the plurality of ad impressions, wherein the adjusting the generated competitive market value prediction for the plurality of ad impressions further comprises;

receiving and identifying, by the engine, one or more additional inputs; and

determining, by the engine, adjusted candidate clearing prices based at least in part on the one or more additional inputs and historical clearing prices of one or more ad impressions corresponding to the one or more additional inputs;

dynamically determining, by the engine, a difference between the adjusted competitive market value based at least in part on the one or more additional inputs and the generated competitive market value;

repeatedly adjusting, by the engine, the generated competitive market value prediction if the determined difference is above a predetermined threshold level; and

finalizing the competitive market value prediction for the plurality of ad impressions on the plurality of ad impression exchanges if the determined difference is below a predetermined threshold level,

wherein one of the finalized competitive market value prediction, the aggressiveness value, and the one or more targets is used to determine a plurality of fair market value bids for the plurality of ad impressions for the advertiser,

wherein the computer system comprises a computer processor and an electronic storage medium.

2. The computer-implemented method of claim 1, wherein the selecting the subset of the candidate set of inputs further comprises defining an initial period and portion of an ad campaign budget for limiting the subset of the candidate set of inputs and the historical clearing prices for the determination of the candidate clearing prices.

3. The computer-implemented method of claim 1, wherein the plurality of fair market value bids is further determined based on at least one of the following types of goals:

cost per click (CPC), cost per lead (CPL), cost per sale (CPS), cost per action (CPA), return on investment (ROI), or winning a targeted portion of available ad impressions.

4. The computer-implemented method of claim 1, wherein the plurality of fair market value bids is determined in a context of a specific ad campaign.

9. The computer-implemented method of claim 1, wherein one or more of the plurality of fair market value bids is used to place a bid for one of the plurality of ad impressions.

11. The computer-implemented method of claim 1, wherein one of the plurality of the ad impression exchanges uses a different communication protocol from another one of the plurality of ad impression exchanges.

5. A system for determining competitive market value predictions for a plurality of ad impressions across a plurality of ad impression exchanges, the system comprising:

one or more computer readable storage devices configured to store a plurality of computer executable instructions; and

one or more hardware computer processors in connection with the one or more computer readable storage devices and configured to execute the plurality of computer executable instructions in order to cause the system to generate a competitive market value prediction for the plurality of ad impressions on the plurality of ad impression exchanges, wherein the system is further caused to;

receive a candidate set of inputs from an advertiser defining two or more criteria for selecting the plurality of ad impressions;

receive an aggressiveness value representative of a level of aggressiveness of the advertiser for bidding on the plurality of ad impressions;

receive one or more targets desired by the advertiser for the plurality of ad impressions, wherein the one or more targets comprise one or more of a targeting segment, channel, or daypart;

dynamically select a subset of the candidate set of inputs for determining candidate clearing prices for the plurality of ad impressions on the plurality of ad impression exchanges, wherein the dynamically selecting the subset of the candidate set of inputs is based at least in part on a profile of one or more past successful bids, wherein the candidate set of inputs comprises two or more selected from a group consisting of;

a campaign goal of the advertiser, a budget of an ad campaign of the advertiser, performance of an ad campaign of the advertiser, or a spending pace for the budget;

receive data relating to historical clearing prices for the plurality of ad impressions on the plurality of ad impression exchanges, each of the historical clearing prices comprising an ad impression bid price that has been accepted on the plurality of ad impression exchanges;

determine the candidate clearing prices, wherein to determine the candidate clearing prices the system is further caused to;

perform normalization of data transmission interfaces of each of the plurality of ad impression exchanges by;

accessing the data transmission interfaces of each of the plurality of ad impression exchanges; and

translating data relating to the candidate set of inputs and the historical clearing prices among the data transmission interfaces of each of the plurality of ad impression exchanges into the single interface; and

select a subset of the historical clearing prices, wherein the selecting is performed by using the selected subset of the candidate set of inputs, wherein the selected subset of the historical clearing prices correspond to the candidate clearing prices, wherein each of the candidate clearing prices is for an ad impression corresponding to the selected subset of inputs;

generate, based at least in part on the candidate clearing prices, the competitive market value prediction for the plurality of ad impressions on the plurality of ad impression exchanges, the competitive market value prediction comprising a distribution function of bid prices and probabilities of each of the bid prices being accepted on the plurality of ad impression exchanges for the plurality of ad impressions;

dynamically adjust the generated competitive market value prediction for the plurality of ad impressions by;

receiving and identifying one or more additional inputs; and

determining adjusted candidate clearing prices based at least in part on the one or more additional inputs and historical clearing prices of one or more ad impressions corresponding to the one or more additional inputs;

dynamically determine a difference between the adjusted competitive market value based at least in part on the one or more additional inputs and the generated competitive market value;

repeatedly adjust the generated competitive market value prediction if the determined difference is above a predetermined threshold level; and

finalize the competitive market value prediction for the plurality of ad impressions on the plurality of ad impression exchanges if the determined difference is below a predetermined threshold level,

wherein one of the finalized competitive market value prediction, the aggressiveness value, and the one or more targets is used to determine a plurality of fair market value bids for the plurality of ad impressions for the advertiser.

6. The system of claim 5, wherein the plurality of computer executable instructions are configured to further cause the system to define an initial period and portion of an ad campaign budget for limiting the subset of the candidate set of inputs and the historical clearing prices for the determination of the candidate clearing prices.

7. The system of claim 5, wherein the plurality of fair market value bids is further determined based on at least one of the following types of goals:

cost per click (CPC), cost per lead (CPL), cost per sale (CPS), cost per action (CPA), return on investment (ROI), or winning a targeted portion of available ad impressions.

8. The system of claim 5, wherein the plurality of fair market value bids is determined in a context of a specific ad campaign.

10. The system of claim 5, wherein one or more of the plurality of fair market value bids is used to place a bid for one of the plurality of ad impressions.

12. The system of claim 5, wherein one of the plurality of the ad impression exchanges uses a different communication protocol from another one of the plurality of ad impression exchanges.