Risk and fraud decisioning isn’t perfect, if it was, the process of identifying which consumers you should do business with would be simple. A particular challenge is the age-old concept of the ‘thin file’ – individuals for whom you don’t have any history or who have ‘fallen through the cracks’. Until now, data vendors have struggled with this population. However, things are changing.
New research from credit reporting company Experian has revealed that there are over 5 million people in the UK who are virtually invisible to the financial system, because there is insufficient information available about their financial track record.1
In the absence of sufficient information to make a decision, this ‘invisible population’ or ‘credit invisibles’ can often either be excluded from mainstream financial services or be forced to pay premium rates on products such as loans, credit cards and mortgages. This is usually because, without any historical credit information, lenders will increase product pricing to offset the perceived increase in risk.
However, there are two distinct areas where great progress has been made to overcome this issue. The first, perhaps obviously, is in the use of a greater variety of data sources - the Experian report references rental, utilities and open banking data to start to fill the gaps. This will certainly help but will it go far enough? Other non-financial datasets hold the key to providing a fuller view, from social media profiles to device information. As machine learning and AI decisioning develops, the mass of data from these channels will be de-crypted and fed into the process, clearly helping decisioning in specific demographics.
What this demonstrates is that risk and fraud decisioning should no longer rely wholly on credit reference agency data. This brings us onto the other area of development – multi-source decisioning. The three UK consumer credit reference agencies hold broadly similar data – similar but different. You could expend a huge amount of energy trying to work out how and where they differ, but the key is in accepting that they do.
Thanks to improved connectivity via API gateways that reach out and return data from multiple sources in the blink of an eye, businesses can now make stronger, more inclusive decisions by taking in a range of information from different vendors. Data costs can be kept under control through rule sets that decide in what order and circumstances different sources are contacted.
This concept is not limited to credit decisioning – payments too can benefit from a gateway and platform that take a data-agnostic approach to validation and decisioning. The Optimize platform that we’ve developed at Pay360 is a case in point. We’ve integrated with 65 data vendors which feed into our decisioning platform to help businesses increase customer acceptance rates and protect their revenue from fraud, providing a friction-free customer journey along the way.
This type of fraud engine can be really powerful in terms of impact – for example, in the first year of a well-known transport business going live with Optimize, there was an increase in prevented fraud of almost £4 million pounds.
It’s clear that a joined-up approach to data and subsequent decisioning can have distinct benefits for business, but what it also shows is that there are many positive outcomes for consumers, not just through a better customer experience, but in driving greater social inclusivity for those currently underserved in the finance world.