Having worked in fraud prevention operations within the banking industry for over 10 years, Sean Neary joined Featurespace as a Subject Matter Expert in Financial Services in December 2016. As 2017 draws nearer, Sean shares his predictions for fraud trends that will be hitting financial services organisations in 2017 – and recommends the steps that organisations can take to protect themselves using the latest machine learning technology.

 

How has fraud evolved in 2016?

For most of this year, I was managing fraud systems within banking and payments, and have seen fraud attacks continue to rise in type and frequency. I’ve seen first-hand that criminals are advancing their methods and sophistication at a rate that is difficult for many existing fraud systems to keep up with.

One example of a type of fraud attack I’ve seen increasing within financial services, is where criminals manipulate the standard authorisation message. For example, to push systems into Stand In Processing (STIP) – where most organisations are limited on their protection capabilities – or to mislead the rule-base into thinking it was a secure transaction.

These attacks impact payment processors upstream of where fraud systems usually spot an attack at the transaction stage.

 

What fraud trends can we expect to see in 2017?

  • Authorisation Stream fraud attacks will dramatically increase as criminals look for fraud opportunities throughout the entire Authorisation Stream, including attacks on the host/processing systems.
  • Social Engineering attacks are also an increasing threat – particularly for vulnerable and elderly banking customers, who are manipulated by a criminal impersonating the bank via phone or email.
  • Risk of increasing fraud operations costs because existing fraud systems cannot take the strain of these new threats, which results in heavy recruitment into fraud operation teams to try and keep up with spotting fraud.

To block these fraud attacks, financial services organisations need to be identifying anomalies quickly, accurately and efficiently at the level of accounts, merchants, individual cardholders and multiple locations, as well as holistically within their host/processing systems.

So, what should businesses do? Well, time for some good news. New machine learning systems – which use adaptive behavioural analytics to monitor individuals in real time and detect anomalies – are enabling financial services organisations to understand behaviour across their customer base, and automatically spot and block new fraud attacks as they occur.

 

How should financial services prioritise fraud protection in 2017?

With this in mind, I see four areas where financial services organisations can take steps to protect themselves better from fraud attacks in 2017.

  • ADAPT – why not be your company champion for adapting to true end-to-end fraud protection? Financial Services organisations are typically already capturing all the data they need to protect their customers throughout the authorisation stream and transaction process – but they need the right advanced anomaly detection fraud system to do it.
  • FOCUS on finding automated ways to fill the gaps which currently make your customers vulnerable to fraud. For example, areas where only static Stand In procedures are used – that were either outsourced to another system and forgotten about, or even left idle since the original system was built.
  • EMBRACE machine learning. Move away from poorly-structured APIs (Application Programmable Interfaces) and systems which solely rely on rule-base models or consortium data.  Fraud trends – like those outlined above – are moving too fast for these systems to block. A machine learning platform such as ARIC, which self-learns as new types of fraud are identified, automatically understands behavior to block fraud as it occurs. At the same time, it understands customer behaviour to dramatically reduce the number of genuine transactions declined (also known as ‘false positives’) – reducing customer friction.
  • BE PROACTIVE – why wait for incidents to occur before finding the point of compromise? Be the fraud prevention hero in your company: make 2017 the year you implement real-time adaptive behavioural analytics to automatically alert you when an anomaly in behaviour is identified.

So, give yourself a break this festive season – and then make 2017 the year you embrace machine learning to tackle fraud prevention.

 

Key takeaways

  • Fraud attacks across the entire Authorisation Stream are expected to rise in 2017 – as are social engineering attacks on the elderly and vulnerable.
  • The machine learning, adaptive behavioural analytics technology already exists to help Financial Services organisations tackle these problems in 2017.
  • 2017 should be the year Financial Services organisations proactively adapt their end-to-end fraud prevention by adopting one accurate, machine learning system to protect their customers and their revenue.