Cross-screen optimization of advertising placement
The current invention relates to a computer-generated method for optimizing placement of advertising content across multiple different devices. The system can allocate advertising campaigns and plans to various inventory types based on the probability of accurate consumer matching. Consumer matching can be achieved by generation of look-alike models in a consumer's device graph to predict future consumption behavior. The system includes an interface through which an advertiser can access relevant information about inventory and success of a given placement.
1. A method, comprising:
defining a pool of consumers based on a consumer graph of consumer properties, the consumer graph comprising a plurality of consumer nodes, the consumer graph being associated with a device graph;
causing an item of advertising content to be delivered to a first device of a consumer in the pool of consumers with an expectation to receive a message related to the advertising content from the first device of the consumer, wherein an identity of the consumer is masked using a consumer ID, the consumer ID being associated with a particular node for the consumer in the consumer graph, the first device of the consumer being selected from the device graph based on the consumer ID;
receiving data indicative of an event of the consumer from a second client device instead of from the first device of the consumer;
determining that the second client device belongs to the consumer based on receiving the data indicative of the event of the consumer from the second client device instead of from the first device of the consumer; and
associating the second client device with the consumer ID in the device graph.
2. The method of claim 1 , wherein the item of advertising content to be delivered to a first device in view of an advertising campaign, the advertising campaign including a certainty estimate for reaching a desired audience, wherein the advertising campaign is adjusted based on an actual performance measured against an estimated certainty of reaching the desired audience.
3. The method of claim 2 , wherein the actual performance includes the event of the consumer.
4. The method of claim 2 , wherein the advertising campaign is adjusted before a conclusion of the advertising campaign based on a correlation of the event to the advertising campaign.
5. The method of claim 4 , wherein the advertising campaign is adjusted before by implementing a machine learning process to improve the estimated certainty of reaching the desired audience.
6. The method of claim 1 , wherein the event is inferred based on location data of the first device, wherein the location data of the first device is indicative of the event.
7. The method of claim 6 , wherein the item of advertising content is related to a particular merchant, wherein the location data of the first device indicates that the consumer likely visited a physical store of the particular merchant.
8. The method of claim 1 , wherein the second client device is determined to belong to the consumer based on a deterministic data of the consumer.
9. The method of claim 1 , wherein the second client device is determined to belong to the consumer based on a probabilistic inference using one or more attributes of the consumer.
10. The method of claim 2 further comprising measuring an effectiveness of the advertising campaign by connecting consumer activity with anonymized advertisement exposure data from a plurality of consumers and using at least one of: machine learning or attribution modeling to measure the effectiveness of the advertising campaign.
11. The method of claim 2 , wherein the advertising campaign is adjusted based on advertisement exposure to the pool of consumers.
12. The method of claim 3 , wherein the advertising campaign is adjusted substantially in real-time.
13. The method of claim 3 , wherein the advertising campaign is adjusted based on a lift analysis.
14. The method of claim 3 , wherein the advertising campaign is adjusted based on advertisement exposure data and consumer behavior.
15. The method of claim 2 , wherein the item of advertising content is to be delivered to the first device of the consumer in a particular slot that has been designated for advertising content.
16. The method of claim 15 , the slot including an exclusion to prevent an entity from obtaining a second slot within a certain boundary of the particular slot.
17. A non-transitory computer-readable storage media storing thereon instructions that, in response to execution by a one or more processors, causes a system to perform operations comprising:
define a pool of consumers based on a graph of consumer properties and a graph of devices;
cause an item of advertising content to be delivered to a first device of a consumer in the pool of consumers with an expectation to receive a message related to the advertising content from the first device of the consumer, the first device of the consumer being associated with a consumer ID in the graph of devices;
receive data indicative of an event of the consumer from a second client device instead of from the first device of the consumer;
determine that the second client device belongs to the consumer; and
associate the second client device in the graph of devices using the consumer ID.
18. The non-transitory computer-readable storage media of claim 17 , wherein the graph of devices is updated to include the second client device being associated with at least one of the first device or the consumer.