Contact stream optimization considering FEC and CC constraints
AssigneeInternational Business Machines Corporation
InventorsFrancisco Jauffred, Mukul Goyal, Glen Osterhout, Ruby Kennedy, Robert Crites
Life of this patent
Estimating the term…
Abstract
Methods, systems, and software for conducting a marketing campaign are described. One or more scores associated with sending offers to proposed contacts are received; and an overall campaign score corresponding to a sum of the scores is optimized to generate optimal assignments of offers to customers. The optimal assignments are included in the marketing campaign.
Claims27
A computer-implemented method for conducting a marketing campaign comprises:
- receiving by one or more computers scores associated with sending offers to proposed contacts belonging to a pool of customers;
- randomly assigning customers in the pool of customers into predetermined groups of customers;
- and
- optimizing by the one or more computers an overall campaign score corresponding to a sum of the received scores, with optimizing generating optimal assignments of offers to customers that satisfy for-each-customer (FEC) constraints and cross-customer (CC) constraints, the FEC constraints controlling communication to an individual customer and the CC constraints representing resources shared among the pool of customers, the optimal assignments of offers for inclusion in the marketing campaign;
- by:
- determining a first assignment of offers for a first group of customers that maximizes a sum of scores for the first group and complies with the FEC constraints and the CC constraints;
- and
- determining by the one or more computers subsequent assignments of offers for the remaining groups using information associated with determining the first assignment, the subsequent assignments maximizing the sums of scores for the remaining groups and complying with the FEC constraints and the cross-customer constraints, the optimal assignments including the first assignments and the subsequent assignments.
A computer readable medium having stored thereon, software for conducting a marketing campaign, the software comprising instructions for causing one or more processors to:
- receive scores associated with sending offers to proposed contacts belonging to a pool of customers;
- randomly assign customers in the pool of customers into predetermined groups of customers;
- and
- optimize an overall campaign score corresponding to a sum of the received scores, wherein optimizing generates optimal assignments of offers to customers that satisfy for-each-customer (FEC) constraints and cross-customer (CC) constraints, the FEC constraints controlling communication to an individual customer and the CC constraints representing resources shared among the pool of customers, the optimal assignments of offers for inclusion in the marketing campaign, by instructions to:
- determine a first assignment of offers for a first group of customers that maximizes a sum of scores for the first group and complies with the FEC constraints and the CC constraints;
- and
- determine subsequent assignments of offers for the remaining groups using information associated with determining the first assignment, the subsequent assignments maximizing the sums of scores for the remaining groups and complying with the FEC constraints and the cross-customer constraints, the optimal assignments including the first assignments and the subsequent assignments.
A system for conducting a marketing campaign comprises:
- memory configured to store scores associated with sending offers to proposed contacts belonging to a pool of customers;
- and
- one or more processors configured to:
- randomly assign customers in the pool of customers into predetermined groups of customers;
- optimize an overall campaign score corresponding to a sum of the scores, to generate optimal assignments of offers to customers that satisfy for-each-customer (FEC) constraints and cross-customer (CC) constraints, the FEC constraints controlling communication to an individual customer and the CC constraints representing resources shared among the pool of customers, the optimal assignments being included in the marketing campaign;
- determine a first assignment of offers for a first group of customers that maximizes a sum of scores for the first group and complies with the FEC constraints and the CC constraints;
- and
- determine subsequent assignment of offers for remaining groups using information associated with determining the first assignment, the subsequent assignments maximizing the sums of scores for the remaining groups and complying with the FEC constraints and the cross-customer constraints, the optimal assignments including the first assignments and the subsequent assignments.
Dependencies are inferred from the claim text.