System, method, and computer program product for generating a synthetic control group

1. A computer-implemented method comprising:

receiving, with at least one processor, transaction account data of a plurality of transaction accounts in a first time period;

receiving, with the at least one processor via a transaction service provider system, transaction data associated with at least one transaction completed by a first set of transaction accounts of the plurality of transaction accounts with at least one target merchant in a second time period;

generating, with the at least one processor, a synthetic control group comprising a subset of transaction accounts sampled from the first set of transaction accounts;

determining, with the at least one processor using a machine learning model, for each transaction account of the synthetic control group, a propensity score representative of a likelihood of being associated with a test group;

assigning, with the at least one processor based at least partly on the propensity score of each transaction account and historic transaction data, an entropy balancing weight to each transaction account of the synthetic control group;

altering, with the at least one processor based on the transaction data and the synthetic control group, at least one operational parameter of a computer-implemented advertisement program to be executed in a third time period, the at least one operational parameter of the computer-implemented advertisement program comprising at least (i) a number of communications to be transmitted, and (ii) a time of communications to be transmitted; and

executing, with the at least one processor, the computer-implemented advertisement program in the third time period by allocating computing resources at least partly based on the at least one operational parameter, wherein allocating the computing resources comprises determining a proportional amount of processing capacity and memory storage for the number of communications to be transmitted and determining which computing resources to use based on the time of communications to be transmitted in comparison to server uptime.

2. The computer-implemented method of claim 1 , wherein determining the propensity score for each transaction account of the synthetic control group is based on at least one of: amount transacted with merchant; amount transacted for merchant type; amount transacted for transaction type; amount transacted in first time period; amount transacted in second time period; or any combination thereof.

3. The computer-implemented method of claim 1 , wherein the entropy balancing weight assigned to each transaction account of the synthetic control group is further based on a respective propensity score of the transaction account.

4. The computer-implemented method of claim 3 , wherein the propensity score is further determined, with the at least one processor using a machine learning model, for each transaction account of the test group.

5. The computer-implemented method of claim 4 , further comprising determining, with the at least one processor and using a machine learning model, for each transaction account of the synthetic control group and the test group, a predictive spending score for the third time period.

6. The computer-implemented method of claim 5 , wherein the entropy balancing weight assigned to each transaction account of the synthetic control group is further based on a respective predictive spending score of the transaction account.

7. The computer-implemented method of claim 1 , wherein the at least one operational parameter of the computer-implemented advertisement program further comprises a list of addresses of communications to be transmitted.

8. A system comprising a server comprising at least one processor, the server being programmed and/or configured to:

receive transaction account data of a plurality of transaction accounts in a first time period;

receive, via a transaction service provider system, transaction data associated with at least one transaction completed by a first set of transaction accounts of the plurality of transaction accounts with at least one target merchant in a second time period;

generate a synthetic control group comprising a subset of transaction accounts sampled from the first set of transaction accounts;

determine, using a machine learning model, for each transaction account of the synthetic control group, a propensity score representative of a likelihood of being associated with a test group;

assign, based at least partly on the propensity score of each transaction account and historic transaction data, an entropy balancing weight to each transaction account of the synthetic control group;

alter, based on the transaction data and the synthetic control group, at least one operational parameter of a computer-implemented advertisement program to be executed in a third time period, the at least one operational parameter of the computer-implemented advertisement program comprising at least (i) a number of communications to be transmitted, and (ii) a time of communications to be transmitted; and

execute the computer-implemented advertisement program in the third time period by allocating computing resources at least partly based on the at least one operational parameter, wherein allocating the computing resources comprises determining a proportional amount of processing capacity and memory storage for the number of communications to be transmitted and determining which computing resources to use based on the time of communications to be transmitted in comparison to server uptime.

9. The system of claim 8 , wherein the entropy balancing weight assigned to each transaction account of the synthetic control group is further based on a respective propensity score of the transaction account.

10. The system of claim 9 , wherein the propensity score is further determined, using the machine learning model, for each transaction account of the test group.

11. The system of claim 10 , wherein the server is further programmed and/or configured to determine, using the machine learning model, for each transaction account of the synthetic control group and the test group, a predictive spending score for the third time period.

12. The system of claim 11 , wherein the entropy balancing weight assigned to each transaction account of the synthetic control group is further based on a respective predictive spending score of the transaction account.

13. The system of claim 8 , wherein the at least one operational parameter of the computer-implemented advertisement program further comprises a list of addresses of communications to be transmitted.

14. A computer program product comprising at least one non-transitory computer-readable medium including program instructions that, when executed by at least one processor, cause the at least one processor to:

receive transaction account data of a plurality of transaction accounts in a first time period;

receive, via a transaction service provider system, transaction data associated with at least one transaction completed by a first set of transaction accounts of the plurality of transaction accounts with at least one target merchant in a second time period;

generate a synthetic control group comprising a subset of transaction accounts sampled from the first set of transaction accounts;

determine, using a machine learning model, for each transaction account of the synthetic control group, a propensity score representative of a likelihood of being associated with a test group;

assign, based at least partly on the propensity score of each transaction account and historic transaction data, an entropy balancing weight to each transaction account of the synthetic control group;

alter, based on the transaction data and the synthetic control group, at least one operational parameter of a computer-implemented advertisement program to be executed in a third time period, the at least one operational parameter of the computer-implemented advertisement program comprising at least (i) a number of communications to be transmitted, and (ii) a time of communications to be transmitted; and

execute the computer-implemented advertisement program in the third time period by allocating computing resources at least partly based on the at least one operational parameter, wherein allocating the computing resources comprises determining a proportional amount of processing capacity and memory storage for the number of communications to be transmitted and determining which computing resources to use based on the time of communications to be transmitted in comparison to server uptime.

15. The computer program product of claim 14 , wherein the entropy balancing weight assigned to each transaction account of the synthetic control group is further based on a respective propensity score of the transaction account.

16. The computer program product of claim 15 , wherein the propensity score is further determined, using the machine learning model, for each transaction account of the test group.

17. The computer program product of claim 16 , wherein the program instructions further cause the at least one processor to determine, using the machine learning model, for each transaction account of the synthetic control group and the test group, a predictive spending score for the third time period.

18. The computer program product of claim 17 , wherein the entropy balancing weight assigned to each transaction account of the synthetic control group is further based on a respective predictive spending score of the transaction account.

19. The computer program product of claim 14 , wherein the at least one operational parameter of the computer-implemented advertisement program further comprises a list of addresses of communications to be transmitted.