10 ways to sharpen your approach to payment fraud analytics

0

It can be hard to stay up to date with rapidly moving fields like payment fraud. For those who are short of time, here is a roundup of all you need to know about the state of the art in the field, including blogs, videos, webinars, white papers and even interactive demos.

Sometimes understanding payment fraud means taking a wider look at fraud more generally. In February 2019, the Association of Certified Fraud Examiners (ACFE) surveyed its members to find out more about what organisations were doing – and particularly, what technologies they were using – to combat fraud. The results of the survey have been put into SAS Visual Analytics to create an interactive demo, so you can find out more about what’s happening in the world of anti-fraud technology.

Hidden Insights: Ten insights on payment fraud analytics to sharpen your approach

Ten ways to sharpen your approach to payment fraud analytics.

Payment fraud case studies

One of the best ways to understand how technology can help is to hear about the experience of particular organisations.

A case study describes how payments processor Nets has improved its fraud detection rate by 50%, reduced card fraud by 50% to 70% with an optional prevention program, and cut false positives in half by using SAS Fraud Management software. The system operates in real time to reduce vulnerability and ensure that Nets can protect its clients.

More general information, as well as a case study, is available from a solution brief on reducing payment fraud. It describes how PKO Bank Polski has been able to reduce its losses from payment fraud by improving levels of detection and reducing false positives using SAS technology. It explains the problems faced by banks, including insufficient fraud detection systems, lack of predictive ability, high costs of doing business and complexity, and shows how SAS software can help to address these through data integration, detection and alert generation tools, fraud network analysis, case management and alert triage management.

Payment fraud in a dynamic environment

The payments industry is dynamic and fast paced, with opportunities and challenges for banks and customers alike changing quickly. This useful infographic describes the top five trends in the field of payments to provide context for efforts to reduce payment fraud. They include the rise of digital, and the understanding that it is essential to operate in real time.

Sundeep Tengur’s blog post "Is your organisation doing enough to keep pace with payment fraud?" explains that new payment methods will deliver huge improvements in speed of transaction, mobility and ease of use, but also provide challenges for fraud detection systems. Sundeep suggests that AI-based fraud detection methods may be essential to stay ahead and make better contextual decisions.

This built on an earlier post in which Sundeep explained that new European payment regulations contained in a new Payment Services Directive (PSD2) could elevate fraud risk. The changes to the payments landscape, including new providers and new services, are likely to be hugely beneficial for customers – but will also offer new vulnerabilities and potential for fraud.

A similar theme is discussed in the webinar Digitalisation: Menace or Opportunity for Fraud Detection? Digitalisation is both opportunity and menace: menace because it opens new areas of vulnerability to fraud, but opportunity because advanced analytics and machine learning tools provide much better ways to detect fraud in real time.

Marcin Nadolny, Head of Fraud Practice South EMEA at SAS and fraud specialist for almost 15 years, picks up this theme in a video on the role of analytics in combating application and payment fraud. He talks about online payments, the main reasons for the need to increase protection against fraud, and how analytics and machine learning capabilities can support this cause.

A balancing act

Fraud detection is made more complicated by the need to avoid inconveniencing legitimate customers going about their rightful business. Too many false positives – that is, preventing real transactions from going ahead – can result in a poor customer experience.

In a recent blog post, Manuel Rodriguez discussed what he described as a "machine learning balancing act." He suggested that banks and other payment providers would need to move from rules-based systems to machine learning to combat payment fraud effectively without inconveniencing customers by rejecting legitimate payments.

#Digitalisation is both opportunity and menace: menace because it opens new areas of vulnerability to fraud, but opportunity because #advancedanalytics and #machinelearning tools provide much better ways to detect fraud in real time. Click To Tweet

The white paper Balancing Fraud Detection and the Customer Experience takes this idea further. It suggests that it may be more important to identify the good customers than the fraudulent ones. A modern concept of digital identity coupled with analytics would allow seamless authentication of good customers and detection of fraudulent identities in real time.

Share

About Author

Ulrike Bergmann

As analytics driven decisions become more widespread, organisations are looking to improve the underlying data management, compliance with privacy regulations and agility of their analytics platform. In my role as campaign manager, I research customer needs and wants, align available tools to address these requirements and curate best practices to help improve the effectiveness of analytics for our customers’ business.

Leave A Reply

Back to Top