Intelligent application of reserves to transactions

1 . A system comprising:

one or more processors; and

one or more non-transitory computer-readable media storing instructions executable by the one or more processors, wherein the instructions cause the one or more processors to perform acts comprising:

receiving, from a plurality of point-of-sale (POS) devices of a plurality of merchants, first transaction data associated with a plurality of first transactions, wherein the first transaction data includes, for an individual first transaction of the plurality of first transactions, an indication of (1) whether a chargeback request was associated with the individual first transaction, and (2) an item associated with the individual first transaction and a customer associated with the individual first transaction;

processing, based at least in part on the first transaction data, the plurality of first transactions;

training a predictive model based at least in part on the first transaction data, wherein the predictive model is trained to output a value reflective of a level of risk associated with a transaction;

receiving, from a POS device of a merchant, an indication of an item associated with a second transaction or a customer associated with the second transaction;

determining, based at least in part on inputting to the predictive model at least the item associated with the second transaction or the customer associated with the second transaction, a value reflective of a level of risk associated with the second transaction;

determining, based at least in part on the level of risk associated with the second transaction, to withhold a portion from a settlement amount of the second transaction, at least some of the portion of funds to be remitted to the merchant at a subsequent time; and

causing deposit of the portion in a reserve account to be associated with a payment processing service.

2 . The system as claim 1 recites, wherein the chargeback request comprises a first chargeback request, the acts further comprising:

receiving a second chargeback request associated with the second transaction; and

based at least in part on receiving the second chargeback request, satisfying a cost of a chargeback associated with the second chargeback request using the reserve account.

3 . The system as claim 1 recites, wherein the portion comprises a first portion, the acts further comprising:

determining a second portion to withhold from the settlement amount of the second transaction, wherein the second portion is associated with a service fee paid to the payment processing service; and

causing the settlement amount, less the first portion and the second portion, to be associated with an account of the merchant.

4 . The system as claim 1 recites, the acts further comprising:

based at least in part on the level of risk associated with the second transaction, determining an offer of insurance for the merchant to insure at least the portion of the second transaction against the risk;

sending the offer to the POS device;

receiving an indication of an acceptance of the offer; and

modifying at least one term of settlement of the second transaction based at least in part on the acceptance of the offer.

5 . The system as claim 1 recites, wherein training the predictive model is further based at least in part on at least one of, for an individual transaction:

a transaction type, wherein the transaction type comprises a card-not-present transaction or a card-present transaction;

geographic location of the individual transaction;

creditworthiness of a merchant associated with the individual transaction; or

fraud risk associated with the merchant associated with the individual transaction.

6 . The system as claim 4 recites, the acts further comprising:

generating, based on the indication that the merchant accepts the offer, an updated value reflective of the level of risk associated with the second transaction.

7 . A method comprising:

receiving, by one or more servers of a payment processing service and from a plurality of point-of-sale (POS) devices of a plurality of merchants, first transaction data associated with a plurality of first transactions, wherein the first transaction data includes, for an individual first transaction of the plurality of first transactions, an indication of (1) whether a chargeback request was associated with the individual first transaction, and (2) an item associated with the individual first transaction and a customer associated with the individual first transaction;

processing, by the one or more servers, based at least in part on the first transaction data, the plurality of first transactions;

training, by the one or more servers, a predictive model based at least in part on the first transaction data, wherein the predictive model is trained to output a value reflective of a level of risk associated with a transaction;

receiving, by the one or more servers from a POS device of a merchant, an indication of an item associated with a second transaction or a customer associated with the second transaction;

determining, by the one or more servers and based at least in part on inputting to the predictive model at least the item associated with the second transaction or the customer associated with the second transaction, a value reflective of a level of risk associated with the second transaction;

determining, by the one or more servers and based at least in part on the level of risk associated with the second transaction, to withhold a portion from a settlement amount of the second transaction, at least some of the portion of funds to be remitted to the merchant at a subsequent time; and

causing deposit, by the one or more servers, of the portion in a reserve account to be associated with the payment processing service.

8 . The method as claim 7 recites, wherein the chargeback request comprises a first chargeback request, further comprising:

receiving, by the one or more servers, a second chargeback request associated with the second transaction; and

based at least in part on receiving the second chargeback request, satisfying, by the one or more servers, a cost of a chargeback associated with the second chargeback request using the reserve account.

9 . The method as claim 7 recites, wherein the portion comprises a first portion, the method further comprising:

determining, by the one or more servers, a second portion to withhold from the settlement amount of the second transaction, wherein the second portion is associated with a service fee paid to the payment processing service; and

causing, by the one or more servers, the settlement amount, less the first portion and the second portion, to be associated with an account of the merchant.

10 . The method as claim 7 recites, further comprising:

based at least in part on the level of risk associated with the second transaction, determining, by the one or more servers, an offer of insurance for the merchant to insure at least the portion of the second transaction against the risk;

sending, by the one or more servers, the offer to the POS device;

receiving, by the one or more servers, an indication of an acceptance of the offer; and

modifying, by the one or more servers, at least one term of settlement of the second transaction based at least in part on the acceptance of the offer.

11 . The method as claim 7 recites, wherein training the predictive model is further based at least in part on at least one of, for an individual transaction:

a transaction type, wherein the transaction type comprises a card-not-present transaction or a card-present transaction;

geographic location of the individual transaction;

creditworthiness of a merchant associated with the individual transaction; or

fraud risk associated with the merchant associated with the individual transaction.

12 . One or more non-transitory computer-readable media storing instructions executable by one or more processors that, when executed by the one or more processors, cause the one or more processors to perform acts comprising:

receiving, from a plurality of point-of-sale (POS) devices of a plurality of merchants, first transaction data associated with a plurality of first transactions, wherein the first transaction data includes, for an individual first transaction of the plurality of first transactions, an indication of (1) whether a chargeback request was associated with the individual first transaction, and (2) an item associated with the individual first transaction and a customer associated with the individual first transaction;

processing, based at least in part on the first transaction data, the plurality of first transactions;

training a predictive model based at least in part on the first transaction data, wherein the predictive model is trained to output a value reflective of a level of risk associated with a transaction;

receiving, from a POS device of a merchant, an indication of an item associated with a second transaction or a customer associated with the second transaction;

determining, based at least in part on inputting to the predictive model at least the item associated with the second transaction or the customer associated with the second transaction, a value reflective of a level of risk associated with the second transaction;

determining, based at least in part on the level of risk associated with the second transaction, to withhold a portion from a settlement amount of the second transaction, at least some of the portion of funds to be remitted to the merchant at a subsequent time; and

causing deposit of the portion in a reserve account to be associated with a payment processing service.

13 . The one or more non-transitory computer-readable media as claim 12 recites, wherein the chargeback request comprises a first chargeback request, the acts further comprising:

receiving a second chargeback request associated with the second transaction; and

based at least in part on receiving the second chargeback request, satisfying a cost of a chargeback associated with the second chargeback request using the reserve account.

14 . The one or more non-transitory computer-readable media as claim 12 recites, wherein the portion comprises a first portion, the acts further comprising:

determining a second portion to withhold from the settlement amount of the second transaction, wherein the second portion is associated with a service fee paid to the payment processing service; and

causing the settlement amount, less the first portion and the second portion, to be associated with an account of the merchant.

15 . The one or more non-transitory computer-readable media as claim 12 recites, further comprising:

storing the value reflective of a level of risk associated with the second transaction in a merchant profile associated with the merchant.

16 . The one or more non-transitory computer-readable media as claim 12 recites, wherein the value reflective of the level of risk is further based at least in part on a predetermined fraud profile associated with a service account of the merchant.

17 . The one or more non-transitory computer-readable media as claim 12 recites, wherein the value reflective of the level of risk is further based at least in part on merchant data.

18 . The one or more non-transitory computer-readable media as claim 12 recites, wherein the value reflective of the level of risk is associated with a likelihood that the second transaction will be associated with a chargeback for a fraud-related reason.

19 . The one or more non-transitory computer-readable media as claim 12 recites, wherein the value reflective of a level of risk associated with the second transaction is based on inputting to the predictive model the item associated with the second transaction.

20 . The one or more non-transitory computer-readable media as claim 12 recites, wherein the value reflective of a level of risk associated with the second transaction is based on inputting to the predictive model the customer associated with the second transaction.