Managing impressions of an advertisement campaign
The present disclosure provides for management of impressions in advertisement campaigns. Impressions may be moved between different impression media based on performance metrics and historical data. Impression budgets may be modified in an active campaign based on forecasts determined using current performance data. Impression budgets may be reallocated between simultaneously active advertisement campaigns.
1. A method for managing impressions comprising:
identifying a first platform and a second platform on which an advertisement is provided, wherein the advertisement has allocated a respective number of budgeted impressions;
receiving, from the first platform via a network, a first real-time popularity of a subject of the advertisement on the first platform, a first anticipated spike of popularity of the first platform, and a first historical impression generation data indicative of other advertisements on the first platform that were similar to the advertisement;
receiving, from the second platform via the network, a second real-time popularity of the subject of the advertisement on the second platform, a second anticipated spike of popularity of the second platform, and a second historical impression generation data indicative of other advertisements on the second platform that were similar to the advertisement;
calculating, for the first platform, using processing circuitry, a first likelihood of impression generation for the advertisement based on the first real-time popularity, the first anticipated spike of popularity, and the first historical impression generation data;
calculating, for the second platform, using the processing circuitry, a second likelihood of impression generation for the advertisement based on the second real-time popularity, the second anticipated spike of popularity, and the second historical impression generation data;
comparing, using the processing circuitry, the first likelihood of impression generation and the second likelihood of impression generation;
determining, using the processing circuitry, a high-performing platform of the first platform and the second platform based on the first likelihood or the second likelihood being highest, and a low-performing platform of the first platform and the second platform based on the first likelihood or the second likelihood being lowest;
determining, using the processing circuitry, a redistribution of allocations of the number of budgeted impressions from the low-performing platform to the high-performing platform based on each of the first likelihood and the second likelihood; and
causing to be implemented, using the processing circuitry, the redistribution of allocations from the low-performing platform to the high-performing platform.
2. The method of claim 1 , further comprising:
identifying the other advertisements on the first platform and the other advertisements on the second platform that were similar to the advertisement based on a genre of the subject of the advertisement.
3. The method of claim 1 , wherein an impression is of a first media asset type for the first platform and wherein the impression is of a second media asset type for the second platform.
4. The method of claim 1 , wherein the first historical impression generation data comprises a history of the other advertisements on the first platform.
5. The method of claim 1 , wherein the second historical impression generation data comprises a history of the other advertisements on the second platform.
6. The method of claim 1 , wherein the comparing, using the processing circuitry, the first likelihood of impression generation and the second likelihood of impression generation further comprises:
determining whether the first likelihood of impression generation is greater than the second likelihood of impression generation.
7. The method of claim 6 , wherein the determining whether the first likelihood of impression generation is greater than the second likelihood of impression generation comprises:
computing a likelihood ratio of the first likelihood of impression generation and the second likelihood of impression generation.
8. The method of claim 1 , wherein the first anticipated spike of popularity of the first platform is determined based on future events that may increase impression generation on the first platform.
9. The method of claim 1 , wherein the second anticipated spike of popularity of the second platform is determined based on future events that may increase impression generation on the second platform.
10. A system comprising:
memory;
processing circuitry configured to:
identify a first platform and a second platform on which an advertisement is provided, wherein the advertisement has allocated a respective number of budgeted impressions;
receive, from the first platform via a network, a first real-time popularity of a subject of the advertisement on the first platform, a first anticipated spike of popularity of the first platform, and a first historical impression generation data indicative of other advertisements on the first platform that were similar to the advertisement;
receive, from the second platform via the network, a second real-time popularity of the subject of the advertisement on the second platform, a second anticipated spike of popularity of the second platform, and a second historical impression generation data indicative of other advertisements on the second platform that were similar to the advertisement;
calculate, for the first platform a first likelihood of impression generation for the advertisement based on the first real-time popularity, the first anticipated spike of popularity, and the first historical impression generation data;
calculate, for the second platform a second likelihood of impression generation for the advertisement based on the second real-time popularity, the second anticipated spike of popularity, and the second historical impression generation data;
compare the first likelihood of impression generation and the second likelihood of impression generation;
determine a high-performing platform of the first platform and the second platform based on the first likelihood or the second likelihood being highest, and a low-performing platform of the first platform and the second platform based on the first likelihood or the second likelihood being lowest;
determine a redistribution of allocations of the number of budgeted impressions from the low-performing platform to the high-performing platform based on each of the first likelihood and the second likelihood; and
cause to be implemented the redistribution of allocations from the low-performing platform to the high-performing platform.
11. The system of claim 10 , wherein the processing circuitry is further configured to:
identify the other advertisements on the first platform and the other advertisements on the second platform that were similar to the advertisement based on a genre of the subject of the advertisement.
12. The system of claim 10 , wherein an impression is of a first media asset type for the first platform and wherein the impression is of a second media asset type for the second platform.
13. The system of claim 10 , wherein the first historical impression generation data comprises a history of the other advertisements on the first platform.
14. The system of claim 10 , wherein the second historical impression generation data comprises a history of the other advertisements on the second platform.
15. The system of claim 10 , wherein the processing circuitry is further configured to compare the first likelihood of impression generation and the second likelihood of impression generation by:
determining whether the first likelihood of impression generation is greater than the second likelihood of impression generation.
16. The system of claim 15 , wherein the processing circuitry is further configured to determine whether the first likelihood of impression generation is greater than the second likelihood of impression generation by:
computing a likelihood ratio of the first likelihood of impression generation and the second likelihood of impression generation.
17. The system of claim 10 , wherein the first anticipated spike of popularity of the first platform is determined based on future events that may increase impression generation on the first platform.
18. The system of claim 10 , wherein the second anticipated spike of popularity of the second platform is determined based on future events that may increase impression generation on the second platform.