System and method for media plan generation for a content delivery network
AssigneeTatari, Inc.
InventorsThomas Alexander Hogan, Timothy Davies Morton, Jaeyeong Kim, Michael D. Swinson
Life of this patent
Estimating the term…
Abstract
Media plan generation systems and methods for TV advertising are disclosed. Embodiments of these systems and methods are adapted to generate a media plan comprising a core plan and a test plan. The core plan may be produced based on historical models regarding historical performance of an entity's media while the test plan may be produced based on a predictive model of media performance.
Claims20
A system, comprising:
- a data store maintaining access to a database of media inventory units;
- one or more processors;
- and
- a non-transitory computer-readable medium storing instructions that, when executed by the one or more processors, cause the system to:
- receive a campaign objective, a total budget, and a risk tolerance parameter;
- dynamically partition the total budget into a core budget allocation and a test budget allocation based on the risk tolerance parameter;
- generate a core media segment by selecting a first set of inventory units from the database based on historical performance data associated with an advertiser;
- generate a test media segment by selecting a second set of inventory units from the database, distinct from the first set;
- rank the second set of inventory units using a machine learning model trained to predict a probability of clearance for the inventory units;
- and
- generate a plurality of bids for the test media segment by optimizing the test budget allocation against the ranked second set of inventory units.
A method for generating a media plan, comprising:
- accessing a database of available media inventory units;
- receiving a campaign objective, a total budget, and a risk tolerance parameter;
- dynamically partitioning the total budget into a core budget allocation and a test budget allocation based on the risk tolerance parameter;
- generating a core media segment by selecting a first set of inventory units based on historical performance data associated with an advertiser;
- generating a test media segment by selecting a second set of inventory units, distinct from the first set;
- ranking the second set of inventory units using a machine learning model trained to predict a probability of clearance;
- and
- generating a plurality of bids for the test media segment by optimizing the test budget allocation against the ranked second set of inventory units.
A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to:
- access a database of available media inventory units;
- receive a campaign objective, a budget, and a risk tolerance parameter;
- dynamically partition the budget into a core budget portion and a test budget portion based on the risk tolerance parameter;
- generate a core allocation by selecting a first set of inventory units based on historical performance data associated with an advertiser;
- generate a test allocation by selecting a second set of inventory units, distinct from the first set;
- rank the second set of inventory units using a machine learning model trained to predict a probability of clearance;
- and
- generate a plurality of bids for the test allocation by optimizing the test budget portion against the ranked second set of inventory units.
Dependencies are inferred from the claim text.