Automated fraud detection for point-of-sale devices
Techniques and arrangements for detecting potentially fraudulent transactions made on a point-of-sale (POS) device. In some instances, the POS device may upload a batch of authorization requests corresponding to transactions made on the POS device while the POS device was in an offline mode. A payment service may receive this batch and, as opposed to processing the entire batch, may select a subset for processing. After selecting the subset, the payment service may attempt to authorize payment instruments associated with each transaction of the subset. The payment service may then compare an amount of declined payment instruments to a threshold. If the amount is greater than the threshold, then the payment service may flag the subset and/or the batch as fraudulent and may refrain from processing further transactions of the batch.
1. An automated fraud-detection system configured to designate a group of transactions as potentially fraudulent without analyzing each individual transaction of the group, the system comprising:
one or more processors; and
one or more computer-readable media storing instructions executable by the one or more processors, wherein the instructions program the one or more processors to:
receive, from a device, an indication of the group of transactions;
receive, from the device, a request to process the group of transactions;
select, from the group of transactions, a subset of transactions to process, the subset of transactions being less than the group of transactions;
attempt to authorize, by communicating with one or more computing devices, individual transactions of the subset of transactions;
determine a decline metric based at least in part on a number of transactions associated with the subset of transactions that were declined;
compare the decline metric to a decline threshold;
at least partly in response to determining that the decline metric is less than the decline threshold, select, from the group of transactions, another subset of transactions to process; and
at least partly in response to determining that the decline metric is greater than the decline threshold, refrain from processing additional transactions of the group of transaction.
2. An automated fraud-detection system of claim 1 , wherein, at least partly in response to the determining that the decline metric is less than the decline threshold, the instructions further program the one or more processors to:
attempt to authorize, by communicating with the one or more computing devices, individual transactions of the another subset of transactions;
determine another decline metric based at least in part on a number of transactions associated with the another subset of transactions that were declined; and
compare the another decline metric to the decline threshold.
3. An automated fraud-detection system of claim 1 , wherein, at least partly in response to the determining that the decline metric is greater than the decline threshold, the instructions further program the one or more processors to:
designate the group of transactions as potentially fraudulent; or
flag the group of transactions for subsequent automated fraud analysis.
4. An automated fraud-detection system of claim 1 , wherein:
each transaction of the group of transactions is associated with (i) payment information from a payment instrument and (ii) a cost of the respective transaction; and
attempting to authorize the individual transactions comprises attempting to authorize the payment instrument for the cost of the respective transaction.
5. An automated fraud-detection system of claim 1 , wherein:
the individual transactions are associated with a payment instrument; and
the instructions further program the one or more processors to send, to the device, an indication of which payment instruments from the subset of transactions were authorized and which payment instruments from the subset of transactions were declined.
6. An automated fraud-detection system of claim 1 , wherein the instructions further program the one or more processors to determine a size of the subset of transaction based at least in part on a risk profile of a merchant associated with the device.
7. A method for performing automated fraud-detection comprising:
receiving, from a device, an indication of a group of transactions;
selecting, from the group of transactions, a subset of transactions to process, the subset of transactions being less than the group of transactions;
attempting to authorize, by communicating with one or more computing devices, individual transactions of the sub set of transactions;
determining a decline metric based at least in part on a number of transactions associated with the subset of transactions that were declined;
comparing the decline metric to a decline threshold;
based at least in part on determining that the decline metric is less than the decline threshold, processing at least one additional transaction from the group of transactions; and
based at least in part on determining that the decline metric is greater than the decline threshold, refraining from processing the at least one additional transaction from the group of transaction.
8. A method of claim 7 , wherein, based at least in part on the determining that the decline metric is less than the decline threshold, the method further comprises:
selecting, from the group of transactions, another subset of transactions;
attempting to authorize, by communicating with the one or more computing devices, individual transactions of the another subset of transactions;
determining another decline metric based at least in part on a number of transactions associated with the another subset of transactions that were declined; and
comparing the another decline metric to the decline threshold.
9. A method of claim 7 , wherein, based at least in part on the determining that the decline metric is greater than the decline threshold, the method further comprises:
designating the group of transactions as potentially fraudulent; or
flagging the group of transactions for subsequent automated fraud analysis.
10. A method of claim 7 , wherein:
each transaction of the group of transactions is associated with (i) payment information from a payment instrument and (ii) a cost of the respective transaction; and
the attempting to authorize the individual transactions comprises attempting to authorize the payment instrument for the cost of the respective transaction.
11. A method of claim 7 , wherein:
the individual transactions are associated with a payment instrument; and
the method further comprises sending, to the device, an indication of which payment instruments from the subset of transactions were authorized and which payment instruments from the subset of transactions were declined.
12. A method of claim 7 , the method further comprising determining a size of the subset of transactions based at least in part on a risk profile of a merchant associated with the device.
13. A method of claim 7 , the method further comprising determining the decline threshold based at least in part on a risk profile of a merchant associated with the device.
14. One or more computer-readable media storing computer-executable instructions that, when executed on one or more processors, cause the one or more processors to perform the acts comprising:
receiving, from a device, an indication of a group of transactions;
selecting, from the group of transactions, a subset of transactions to process, the subset of transactions being less than the group of transactions;
attempting to authorize, by communicating with one or more computing devices, individual transactions of the subset of transactions;
determining a decline metric based at least in part on a number of transactions associated with the subset of transactions that were declined;
comparing the decline metric to a decline threshold; and
based at least in part determining that the decline metric is greater than the decline threshold, refraining from processing additional transactions of the group of transactions.
15. One or more computer-readable media as recited in claim 14 , the acts further comprising:
based at least in part determining that the decline metric is greater than the decline threshold:
selecting, from the group of transactions, another subset of transactions to process;
attempting to authorize, by communicating with the one or more computing devices, individual transactions of the another subset of transactions;
determining another decline metric based at least in part on a number of transactions associated with the another subset of transactions that were declined; and
comparing the another decline metric to the decline threshold.
16. One or more computer-readable media as recited in claim 14 , the acts further comprising:
designating the group of transactions as potentially fraudulent; or
flagging the group of transactions for subsequent automated fraud analysis.
17. One or more computer-readable media as recited in claim 14 , wherein:
each transaction of the group of transactions is associated with (i) payment information from a payment instrument and (ii) a cost of the respective transaction; and
the attempting to authorize the individual transactions comprises attempting to authorize the payment instrument for the cost of the respective transaction.
18. One or more computer-readable media as recited in claim 14 , wherein:
the individual transactions are associated with a payment instrument; and
the acts further comprising sending, to the device, an indication of which payment instruments from the subset of transactions were authorized and which payment instruments from the subset of transactions were declined.
19. One or more computer-readable media as recited in claim 14 , the acts further comprising determining a size of the subset of transactions based at least in part on a risk profile of a merchant associated with the device.
20. One or more computer-readable media as recited in claim 14 , the acts further comprising determining the decline threshold based at least in part on a risk profile of a merchant associated with the device.