Helping an EU Bank Holding Company reduce losses and False Positives in Transaction Fraud

CRISL was approached by a large EU-based bank holding company that was plagued by check credit losses. Fraudsters were using fictitious checks and withdrawing funds, and the blocking of funds to control withdrawals was impacting genuine customers. High volumes made manual operations infeasible.

The objective of the engagement was to reduce losses and minimize impact on other customers by designing and implementing a new system for transaction fraud monitoring.

The CRISIL team devised a new methodology that ranks credit entries based on transactional, behavioral and demographical parameters, differentiating high-risk entries from low-risk ones.

Model Implementation

Model Performance

First Name

Last name

Organisation

Designation

Business E-mail

Phone Number

Pin Code

Enter Captcha

I agree to terms and conditions

I agree for CRISIL to communicate with me for useful information

CRISL was approached by a large EU-based bank holding company that was plagued by check credit losses. Fraudsters were using fictitious checks and withdrawing funds, and the blocking of funds to control withdrawals was impacting genuine customers. High volumes made manual operations infeasible.

The objective of the engagement was to reduce losses and minimize impact on other customers by designing and implementing a new system for transaction fraud monitoring.

The CRISIL team devised a new methodology that ranks credit entries based on transactional, behavioral and demographical parameters, differentiating high-risk entries from low-risk ones.

Model Implementation

Model Performance

CRISL was approached by a large EU-based bank holding company that was plagued by check credit losses. Fraudsters were using fictitious checks and withdrawing funds, and the blocking of funds to control withdrawals was impacting genuine customers. High volumes made manual operations infeasible.

The objective of the engagement was to reduce losses and minimize impact on other customers by designing and implementing a new system for transaction fraud monitoring.

The CRISIL team devised a new methodology that ranks credit entries based on transactional, behavioral and demographical parameters, differentiating high-risk entries from low-risk ones.

Model Implementation

Model Performance