Intelligent fraud detection
Rainbird was approached by an international credit card provider that processes over half a million transactions per minute.
The number of reported fraud cases in the financial sector nearly doubled in the last year, with businesses struggling to keep up with ever-shifting vectors of attack, from new modes of mobile fraud to skyrocketing rates of identity fraud.
They wanted to improve fraud detection, reduce the time it takes to deal with false positives and improve customer satisfaction.
The ISMG reported that only 34% of C-level leaders have high confidence in their organisation’s ability to detect and prevent fraud.
The complicated fraud-detection process required an offshore team to continuously monitor data feeds on multiple screens to determine whether customer transactions flagged by their algorithms were fraudulent or genuine. This manual process led to erroneous judgements, resulting in many unnecessarily blocked cards and huge frustration for customers.
40% of cardholders abandon cards after false declines, and a quarter of these people move their cards to the back of their wallets
In 20% of cases, the offshore team phoned the customer directly to request additional information, which resulted in expensive security steps, language barriers, and, in some cases, lost customers.
We spent time with the company’s fraud-prevention Subject Matter Experts (SMEs) to create a Rainbird model that identifies fraud by replicating their best-practice methods, without relying on humans. This reduced the time spent on each case, and improved accuracy and consistent decision-making.
To do this, Rainbird took the logic from over 50 fraud cases, along with the expertise of three of the business’ best-performing SMEs to model a Rainbird-powered fraud engine that represented the company’s best-practice procedures, which was capable of handling multiple concurrent transactions simultaneously and consistently.
We also created a new model for prioritising cases that needed expert involvement so that the most high-risk cases can be assessed by the client’s most experienced on-shore team members.
Rainbird automatically calculates every transaction with a percentage likelihood of being genuine or fraud.
The result of our work was an 85% automation of all cases and a 60% reduction in back-office processing costs.
The new process avoids the less accurate and inconsistent judgements that were being made by the client’s offshore team. The company benefits from:
- Increased detection rates
- A reduction in the number of false positives
- Cost and inconvenience of outbound phone calls to customers
Calls to customers are now backed by Rainbird’s in-built Audit trail, which can explain why Rainbird has concluded a fraud risk and better inform the customer conversation leading to consistent, better-quality calls, and demonstrably better customer outcomes.