Payment fraud risk review management system
The risk team evaluates payment alerts and assigns the necessary checks. Research results help monitor losses and cases where checks unreasonably hinder a customer.
Identity, device, behavioural and transaction data are assessed separately. Fraud is detected too late, or too many legitimate transactions are blocked.
How the solution works
- Receive the signals available at decision time
- Evaluate the risk of a specific fraud scenario
- Return the Permissible Action to the Payment Process
- Investigating an unclear or disputed case
- Check the actual result and the effect of the rule
Key challenges
- Fraud signals not connected in real time
Solution capabilities
Signal Time
For each signal, the time and reliability of reception are kept. The facts revealed after the operation are not used to pretend that they could have been known in advance.
The Script Rule
Separates the signs of stolen access, unusual operation, or trader risk. The total score does not change the reason required for the action.
Permissible action
Additional verification, suspension or transfer for evaluation is selected according to the role and rules of the provider. A low risk score does not in itself exempt from mandatory authentication.
History of the study
The employee sees evidence, the original decision and the information subsequently obtained; completing the investigation does not in itself mean that there was no fraud.
Monitoring results
The loss, funds returned and the wrongly rejected legal payment are assessed separately. Changes to the rules are checked by segment and sufficiently mature data of the results.
Business context
- Fraud signals not connected in real time
- Identity, device, behavioural and transaction data are assessed separately. Fraud is detected too late, or too many legitimate transactions are blocked.
- Risk control is assessed in conjunction with smooth purchasing
- Too strict or inaccurate verification can stop legal settlement and the sale of the trader. The results of the research allow to assess which signals are justified and which cause unnecessary obstacles. The review of the rules is based on data from losses and exposure caused to the customer, while maintaining the decision of the responsible specialists.
Core features
- Signal Time
- The Script Rule
- Permissible action
- History of the study
- Monitoring results
Key integrations
- Payment and authentication processes
- Operation data, available signals and action actually performed.
- Sources of Disputes and Financial Results
- Verified operation result, confirmed loss and recoveries.
Potential impact (%)
The ranges indicate an illustrative relative change in the metric under the stated assumptions. Results depend on the starting position and actual use of the solution. Percentages for different metrics must not be added together.
Lower loss of approved fraud
2–12%Decreasing
This illustrative scenario assumes that the controls described can address 10-30% of discrepancies. That share is assumed to fall by 20-40%. Company data is needed to verify both the addressable share and the resulting change.
The approved net fraud loss in euro is divided by the value of payments in euro of the same group and multiplied by a million.
Number of wrongly rejected legitimate payments
2–12%Decreasing
This illustrative scenario assumes that the controls described can address 10-30% of discrepancies. That share is assumed to fall by 20-40%. Company data is needed to verify both the addressable share and the resulting change.
Confirmed false rejections per thousand operations of similar traffic are counted. Cases not yet solved are shown separately.
Conditional calculation scenarios. The assumptions have not been validated against client measurements.
When this solution is relevant
- Suspicious operations are seen too late when they can no longer be stopped.
- The results of the checks are not linked to later detected frauds, making it difficult to clarify the applicable rules.
Project scope and implementation
Additional checks are combined with measures already taken by the payment provider. Rules and employee decisions are linked to the actual status of the operation and subsequent results of the investigation. Risk rules and payment stop rights are the responsibility of the service provider.
Further development options
- Additional scripts or models only have a sufficient history of their results and the ability to oversee false rejections.