Criminals are exploiting the weak points in the payments process, by using more and more sophisticated technology to attack payment and banking systems, making fraud a significant pain point for financial institutions. Predictions have already reached USD 7 billion in customer-not-present fraud by 2020, for the US alone.
With each new change, such as increasing online payments, chip-enabled cards and contactless technology, criminals focus on the weakest link in the chain. There is a challenge to protect customers from these kinds of new fraud without impacting the customer experience.
This is currently being seen in the US with the introduction of EMV in October 2015, and a huge 215% rise in online fraud in the last 12 months, according to the Global Fraud Index.
Faced with these threats, how can financial institutions stay ahead of the criminals whilst maintaining frictionless customer experience?
The answer
This is where new fraud detection systems using advanced, real-time, deep machine learning technology, stay ahead of the curve. Increasingly, financial institutions and payment processors are looking at deep machine learning methods for their fraud protection. These methods automate the analysis of complex, varied streams of data, making it easier for companies to make faster, more accurate decisions about an individual customer, compared to relying on just human analysis alone.
Featurespace’s unique approach is called Adaptive Behavioural Analytics. Compared to rules-based systems which rely on pattern-matching against known fraud types, Featurespace’s ARIC engine uses this new approach to build up profiles of ‘normal’ behaviour in real-time, so that it can quickly and easily spot the moment that changes occur, whilst understanding the context of the change. As ARIC is focusing on understanding the customers individually and picking out the anomalous data, it is accurate at spotting new and unknown types of fraud. As shown in the graph above, this leads to a significant increase in detection rate as the daily fraud transactions increases, where an existing system’s detection rate declines.
When ARIC detects an anomaly, it provides a risk score based on the known behaviour of the individual customer then, as a self-learning system, the algorithms automatically adapt when a new fraud type is identified, with no need to retune the system.
At the same time, the added benefit of the high accuracy level of Adaptive Behavioural Analytics is that a genuine customer’s behaviour is easy to recognise, enabling financial institutions to accept more business, while reducing customer friction.
The benefits
As the image below illustrates, Featurespace’s ARIC engine uses numerous input data to stop a number of different threats, all in real time. The benefits that our clients are seeing include:
Spotting new fraud attacks at the moment they occur;
70% reduction in genuine transactions declined – enabling financial institutions to accept; more business while minimising customer friction;
54% reduction in chargebacks, reducing fraud losses;
50% improvement in operational efficiency, enabling fraud analysts to focus on the most urgent fraud cases.
Adaptive Behavioural Analytics is transforming the way that organisations in banking, payments, gaming and insurance can prevent fraud while accepting more business from their genuine customers. ARIC’s ability to enhance decision making capabilities around fraud is why TSYS, the largest payments processor in the United States, chose to work with Featurespace to strengthen its position in faster payments using machine learning, to provide clients with actionable insights in real-time.
Andrew Mathieson, Group Executive, issuer product group at TSYS has said: “TSYS’ collaboration with Featurespace aligns with our overall strategy of integrating with advanced, innovative technology partners to help our clients grow their business, reduce costs, and deliver an exceptional customer experience, we will incorporate these capabilities across the credit risk lifecycle, enabling our issuers to catch more fraudulent transactions while dramatically reducing false-positive alerts for genuine transactions — a sharp contrast to the industry paradigm of blocking more valid transactions in order to detect actual fraudulent activity.”
To find out more about Adaptive Behavioural Analytics and the ARIC engine, please visit featurespace.co.uk.
About Luke Reynolds
Luke is Fraud Director at Featurespace and is responsible for Featurespace’s clients in Financial Services and Insurance. Luke has worked in the Financial Services Sector for over 20 years. His previous roles include Commercial Director of Fraud and ID at Callcredit, as well as a variety of positions at Lloyds Banking Group, including Head of Retail Audit, Head of Fraud and Head of Group Security and Investigation.
About Featurespace
Featurespace is the world-leader in Adaptive Behavioural Analytics and creator of the ARIC engine, a machine learning software platform developed out of the University of Cambridge, which understands individual behaviours in real-time for enhanced fraud detection decision-making capabilities. ARIC is deployed in the UK, US and Europe to organisations that have services or products deployed in over 180 countries. Customers include TSYS, Betfair, KPMG, Vocalink/Zapp, Camelot, and William Hill.
Every day we send out a free e-mail with the most important headlines of the last 24 hours.
Subscribe now