Fraud Management Business Case

Increase online transaction security for Banking & Finance, Payment Processors and Insurance with real-time event analysis & CI

Industry

Banking, Finance, Payments Processors, Merchant Acquirers, Insurance Companies and Gaming Organizations.

Objective

To intercept in real time suspicious transactions and prevent online fraud.

Background

The Covid19 crisis has increased the number of transactions for online payments. Customers all around the world were forced to use online stores for goods and services, so the exposure to frauds has progressively increased.

Phishing, scam eCommerce website, and an overall more frequent sharing of personal and payment data – i.e. credit/debit cards details – are jeopardizing security. For companies, this means increased costs for refunds and lower customer loyalty.

Given the real-time nature of online payments, Fraud Detection must follow accordingly.

Business Case

In this scenario, the ideal procedure is to

Integrate data sources in order to analyze data in real-time
Integrate into modern cloud-based services for data visualization and reporting
Analyze transactions in real-time to detect frauds and send notifications
Perform potential fraud analysis to identify patterns and power machine learning algorithms.

This can be achieved via two different approaches:

Real time analysis of events

Real-time analysis of events is based on gathering information from different sources (i.e. credit card payments, geo location of customers app, etc.) and rising an alert when complex correlated events happen. For example, a British customer withdraws cash from an ATM located in UK, and their last mobile banking app login happened abroad; a 70-year-old customer buys a skateboard online, spending 200£ while their average ecommerce purchase is 50£.

In both examples, the alert triggered by the anomalous event demonstrates that applying Event-Driven Architecture (EDA) to real time events can greatly reduce time to reaction and preempt fraud.

Continuous Intelligence

Continuous Intelligence is the application of AI/ML to a stream of events using Event-Driven Architecture. In this way the system auto-learns from standard and non-standard patterns of events, training itself with new, possible scenarios.

In case of anomalous events, this self-learning algorithm triggers alerts and forecasts their impact on the scenario, analyzing past events and patterns of correlation.

For example, the system can notice that small transactions from specific areas of the world are statistically followed by larger transactions with stolen credit cards. In this way, it can preemptively freeze a card or account when suspicious activities are detected.

Fraud Management Business Case

Industry

Objective

Background

Business Case

Real time analysis of events

Continuous Intelligence

Do you want to know more about our services? Fill in the form and schedule a meeting with our team!

Milano

Treviso

Trieste

Chiasso