Predicting advertisement impact for audience selection
An influence system for predicting advertisement impact for audience selection. An advertising probe campaign is operated by sending an advertisement to each entity in a treatment group of entities. A control group of entities which excludes the treatment group entities is selected and no campaign advertising content is sent to the treatment group entities. An influence model is created by comparing features of the treatment group converters to features of the control group converters. An individual frequency cap is selected for each entity that is a candidate for the advertising campaign based on a result of applying the influence model to the features of the candidate entity. The entity may be selected to receive an advertisement based on the individual frequency cap. Some embodiments are integrated with a real time bidding (RTB) exchange and a bid response may be configured based on the results of applying the influence model.
1. A method, comprising:
operating, by an influence system, an advertising probe campaign by:
sending an item of advertising probe campaign content over a network to each entity in a treatment group of entities; and
selecting a control group of entities, wherein the control group excludes entities of the treatment group and the item of advertising probe campaign content is not sent to the entities of the control group; and
assessing results of operating the advertising probe campaign by:
selecting treatment group converters from the treatment group of entities, the treatment group converters having met a conversion criterion of the advertising probe campaign after exposure to the item of campaign advertising content; and
selecting control group converters from the control group of entities, the control group converters having met the conversion criterion of the advertising probe campaign without exposure to the item of campaign advertising content;
constructing an influence model, by the influence system, by comparing features of media consumption histories of the treatment group converters to features of media consumption histories of the control group converters;
generating a blacklist, by the influence system, comprising:
applying the constructed influence model to features of a specified entity's media consumption history; and
assigning the specified entity to a blacklist based on a result of applying the constructed influence model, the blacklist specifying a set of entities which should be excluded from receiving content associated with an advertising campaign; and
operating the advertising campaign, by the influence system, comprising
receiving, by the influence system and from an auction system, a notification that an opportunity to expose the specified entity to advertising content is available;
determining a response to the notification based on the specified entity being on the blacklist; and
sending, from the influence system to the auction system, the response to the notification.
2. The method of claim 1 , wherein:
assigning the specified entity to the blacklist further comprises assigning an individual frequency cap to the specified entity based on the result of applying the constructed influence model; and
determining the response to the notification is further based on the individual frequency cap assigned to the specified entity.
3. The method of claim 2 , wherein the individual frequency cap assigned to the specified entity is zero.
4. The method of claim 1 , wherein:
assigning the specified entity to the blacklist further comprises assigning an individual bid amount to the specified entity based on the result of applying the constructed influence model; and
determining the response to the notification is further based on the assigned bid amount.
5. The method of claim 4 , wherein the assigned bid amount is zero.
6. The method of claim 1 , wherein:
assigning the specified entity to the blacklist further comprises selecting a first advertising content based on the result of applying the constructed influence model;
the bid response further comprises instructions to provide the first advertising content to the specified entity.
7. A non-transitory computer-readable storage medium storing processor-executable computer program instructions that, when executed, cause a computer processor to perform a method, the method comprising:
operating, by an influence system, an advertising probe campaign by:
sending an item of advertising probe campaign content over a network to each entity in a treatment group of entities; and
selecting a control group of entities, wherein the control group excludes entities of the treatment group and the item of of advertising probe campaign content is not sent to the entities of the control group; and
assessing results of operating the advertising probe campaign by:
selecting treatment group converters from the treatment group of entities, the treatment group converters having met a conversion criterion of the advertising probe campaign after exposure to the item of campaign advertising content; and
selecting control group converters from the control group of entities, the control group converters having met the conversion criterion of the advertising probe campaign without exposure to the item of campaign advertising content;
constructing an influence model, by the influence system, by comparing features of media consumption histories of the treatment group converters to features of media consumption histories of the control group converters;
generating a blacklist, by the influence system, comprising:
applying the constructed influence model to features of a specified entity's media consumption history; and
assigning the specified entity to a blacklist based on a result of applying the constructed influence model, the blacklist specifying a set of entities which should be excluded from receiving content associated with an advertising campaign; and
operating the advertising campaign, by the influence system, comprising
receiving, by the influence system and from an auction system, a notification that an opportunity to expose the specified entity to advertising content is available;
determining a response to the notification based on the specified entity being on the blacklist; and
sending, from the influence system to the auction system, the response to the notification.
8. The medium of claim 7 , wherein:
assigning the specified entity to the blacklist further comprises assigning an individual frequency cap to the specified entity based on the result of applying the constructed influence model; and
determining the response to the notification is further based on the individual frequency cap assigned to the specified entity.
9. The medium of claim 8 , wherein the individual frequency cap assigned to the specified entity is zero.
10. The medium of claim 7 , wherein:
assigning the specified entity to the blacklist further comprises assigning an individual bid amount to the specified entity based on the result of applying the constructed influence model; and
determining the response to the notification is further based on the assigned bid amount.
11. The medium of claim 10 , wherein the assigned bid amount is zero.
12. The medium of claim 7 , wherein:
assigning the specified entity to the blacklist further comprises selecting a first advertising content based on the result of applying the constructed influence model;
the bid response further comprises instructions to provide the first advertising content to the specified entity.
13. A system comprising:
a processor; and
a non-transitory computer-readable storage medium storing processor-executable computer program instructions that, when executed, cause a computer processor to perform a method, the method comprising:
operating, by an influence system, an advertising probe campaign by:
sending an item of advertising probe campaign content over a network to each entity in a treatment group of entities; and
selecting a control group of entities, wherein the control group excludes entities of the treatment group and the item of advertising probe campaign content is not sent to the entities of the control group; and
assessing results of operating the advertising probe campaign by:
selecting treatment group converters from the treatment group of entities, the treatment group converters having met a conversion criterion of the advertising probe campaign after exposure to the item of campaign advertising content; and
selecting control group converters from the control group of entities, the control group converters having met the conversion criterion of the advertising probe campaign without exposure to the item of campaign advertising content;
constructing an influence model, by the influence system, by comparing features of media consumption histories of the treatment group converters to features of media consumption histories of the control group converters;
generating a blacklist, by the influence system, comprising:
applying the constructed influence model to features of a specified entity's media consumption history; and
assigning the specified entity to a blacklist based on a result of applying the constructed influence model, the blacklist specifying a set of entities which should be excluded from receiving content associated with an advertising campaign; and
operating the advertising campaign, by the influence system, comprising
receiving, by the influence system and from an auction system, a notification that an opportunity to expose the specified entity to advertising content is available;
determining a response to the notification based on the specified entity being on the blacklist; and
sending, from the influence system to the auction system, the response to the notification.
14. The system of claim 13 , wherein:
assigning the specified entity to the blacklist further comprises assigning an individual frequency cap to the specified entity based on the result of applying the constructed influence model; and
determining the response to the notification is further based on the individual frequency cap assigned to the specified entity.
15. The system of claim 14 , wherein the individual frequency cap assigned to the specified entity is zero.
16. The system of claim 13 , wherein:
assigning the specified entity to the blacklist further comprises assigning an individual bid amount to the specified entity based on the result of applying the constructed influence model; and
determining the response to the notification is further based on the assigned bid amount.
17. The system of claim 16 , wherein the assigned bid amount is zero.
18. The system of claim 13 , wherein:
assigning the specified entity to the blacklist further comprises selecting a first advertising content based on the result of applying the constructed influence model;
the bid response further comprises instructions to provide the first advertising content to the specified entity.