Fraud costs the UK financial services industry £billions a year as fraudsters continue to find ways to outfox fraud detection measures and exploit the proliferation of digitalisation in banking. In this blog, Nick, our Head of Fraud Analysis, explores how lenders can tackle application fraud through data sharing and collaboration with your peers.
Fraud is on the rise. Fortunately, the larger banks typically have adequate volumes of frauds from which to build bespoke models. However, the smaller challenger banks are often constrained due to their size. Fraudsters will often target the smaller and newly established firms in the knowledge of fraud detection being not being as sophisticated as their larger peers.
Why is fraud more of an issue now?
Industry figures indicate there were c30k instances of application fraud during 2017, however this is unlikely to be a full representation of the true picture. Many applications for credit are referred on the basis they are likely to be fraud, though cases remain unresolved due to the applicant being unable to provide the required verification. As lenders have insufficient evidence, or often the time, to prove material falsehoods these applications will be remain in “suspect” queues and not be officially reported as fraud.
There is also a significant element of fraud that is never detected, applications for credit are manipulated to bolster credit scores and secure approval for facilities/rates they would not ordinarily qualify, and this also creates losses.
What we are seeing in the industry
Indeed, the cost of fraud to the industry extends well beyond the losses incurred. Add to this the operational expense of teams and systems deployed to prevent fraud, and the opportunity cost of arising from genuine applications that are abandoned during the inevitable delays in onboarding customers.
However, the industry is doing ever more to share data and case based matching systems continue to grow in number and size. Banks are becoming more sophisticated as they start to use digital footprints to identify suspicious activity. Often, the increase in complexity brings about additional challenges as there is close to an information overload with referrals from a multitude of systems often actioned in silo creating a lengthy inefficient operational process.
How you can tackle fraud now
Here’s three things you can consider to better protect against application fraud now.
1. Up-front data capture
Through intelligent use of the data captured up-front during the initial credit decision, it is possible to create scoring models to identify high and lower risk applications and create bespoke strategies rather than the ‘one size fits all’ approach used by many. This enables the highest levels of scrutiny to be applied to the applicants that are of the highest risk, and conversely the lowest risk cases can be onboarded with minimal delay.
2. Data sharing
Sharing data with peers provides broader insight to build more accurate and robust fraud models – this is particularly useful for smaller lenders who don’t have access to masses of data.
3. Utilising AI technology
Moreover, given the emergence of AI techniques, lenders can more quickly identify suspicious patterns which helps to predict fraud risk with even greater accuracy – helping to avoid associated losses.
With this in mind, Jaywing recently relaunched a fraud detection product (Foil) that enables you to share and gain access to fraud data, for free. In addition, you can choose to opt for its AI technology which we have seen improve fraud detection by a further 10%.