Opinion: Fraud, fintech and definitive action

When it comes to payments, COVID-19 saw the world’s consumers go digital and a decade worth of innovation compressed into the space of several months. And while it seems the worst of the pandemic is behind us, the trend towards digital payments is set to continue. Our Prime-Time for Real-Time report revealed that by 2025, we could see the volume of real-time payments rise to 23 billion payments in Europe. With figures like these, it’s unsurprising many believe we’re at full steam ahead for a cashless society.

But the global trend towards more real-time and digital payments has naturally caught online fraudsters’ attention. In fact, levels of fraud in the UK are now so high it’s a national security threat. According to UK Finance, £754m was stolen from bank customers during the first half of this year - a 30% rise in the same period in 2020

Fraudsters are getting creative

While we’ve seen positive strides in making the online activity more secure, including the introduction of PSD2 and SCA, fraudsters have quickly found ways to target unsuspecting consumers.

This year particularly, scammers focused their activity on authorised push payment (APP) fraud, which increased 71% during the first half of 2021. And with this, in many cases, some scammers have turned to traditional social engineering methods to gather sensitive information, such as the now widespread Post Office text phishing scams.

For financial institutions, it’s hard to keep up. But the latest UK fraud figures must serve as a wake-up call. As an industry it’s vital we find new ways to detect illicit activity if we are to win the battle against fraud. We’re currently emerging from one national crisis, it’s vital we stop sleepwalking into the next.

Collaborating is key

When the pandemic struck, the nation came together by staying apart. Everyone knew their role and understood their personal sacrifices were part of a bigger, more important picture of stopping the spread of the virus. And collaboration didn’t stop there - it extended to big-pharma. All the major healthcare firms, regardless of geographical or commercial loyalties, worked together, sharing intelligence, and cooperating to develop a safe vaccine fit for the masses.

When it comes to fighting the other, very real, and very present pandemic that is digital fraud, it’s time financial institutions take a leaf out of the healthcare industry’s book. From regulators to law enforcement to banks, every party involved in combating and prosecuting fraud must collaborate effectively. But the question is: how?

The solution

One effective option is to create an intelligent network that links organisations together, allowing individual players in the network to build fraud strategies based on a consortium of shared prior knowledge of where and how fraudsters have previously attacked. Ideally, members would connect to the consortium via a central body, such as a central bank, creating a community that shares real-time fraud patterns.

What could this look like in practice? Say there was a high number of transactions made by someone with a specific payment option within a short period of time, such as three hours. This would be flagged as suspicious as most consumers are not usually in a rush to perform that many transactions within a short time frame. When the number of legitimate transactions to fraudulent ones are compared using the consortium data, as the count of transactions within three hours increases, the probability that it is legitimate decreases rapidly.

To harness network intelligence in this way, financial institutions must utilise a hybrid-type Machine Learning (ML) approach. The most effective ones are those that leverage the best intel from every player in a network. A robust ML structure that makes up these networks will enable financial institutions to maintain strong custom-type signals, complemented by signals exchanged in the community, as well as signals from third-party fraud intelligence sources.

Nullifying the virus

The best way to defeat something is to draw attention to it. Just as we did with COVID-19, the more people knew, the more we did to prevent the spread. The industry body of UK finance has said that current fraud levels are now a threat to national security, and people are catching on. But it’s vital we work together, share information, and - ultimately - collaborate to nullify the threat.

Harnessing data-driven insight through network intelligence allows banks, processors, acquirers, and networks to securely share and consume industry-wide fraud signals, so they can feed their ML models alongside proprietary data. This is collaboration in action, and we need more of it. 

About the author: Jackie Barwell is Director of Payments Intelligence at ACI Worldwide. She has spent 25 years in the financial services space (banking), three years in the card payment processing space and more than nine years within the software industry - with a speciality towards fraud risk, prevention, and detection.