"While it is possible to authenticate and risk assess a customer based on few attributes, using numerous, changing and interconnected attributes provides vastly improved KYC and risk certainty."
Whilst mortgage brokers, insurers and other financial intermediaries are not subject to the same rigid anti-money-laundering provisions as financial institutions, it’s nonetheless prudent for all organisations in this field to establish and maintain up-to-date systems and controls to reduce the risks of handling the proceeds of crime, and stay compliant with the Proceeds of Crime Act (2002).
First and foremost, intermediaries must ensure they are maintaining robust KYC processes. Manually verifying customers will only partially reveal the risks associated with an individual and is an increasingly inefficient way of working, with very few data attributes openly accessible for consideration.
Richer data sets can help verify customers faster than ever before, gaining far more detailed insights into each individual. Armed with the latest technological solutions, evaluations can be much more credible, and intermediaries can be safer in the knowledge they’ve done everything within their power to protect against money launderers.
More effective compliance protocols
The advantages of establishing an up-to-date digital customer verification infrastructure are numerous. For one, fast customer authentication improves compliance, and when data science is properly harnessed, it can result in huge cost savings for financial intermediaries. Expanded data sets, coupled with data-led KYC processes not only provide lower-cost KYC compliance with increased certainty of who the customer really is (e.g. what risk they pose, whether a fraudster is impersonating a real person), but they can authenticate in a way that creates considerably less disruption for the customer. Consequently, this makes everyone’s lives easier.
To really make the most of larger data sets and gain the most useful insights, new tools are required. While it is possible to authenticate and risk assess a customer based on few attributes, using numerous, changing and interconnected attributes provides vastly improved KYC and risk certainty. Of course, it also requires increasingly complex analysis, which is beyond the human brain. Enter: machine learning.
A sub-set of artificial intelligence (AI), machine learning, takes large sets of data and finds patterns and connections that can be used to determine identity and associated risk, without the need for the human intervention. Whilst there remains some friction between regulation and the ‘explainability’ of certain AI-driven systems, many data-driven Statistical Modelling and machine learning solutions are available to be deployed now, and are fully compliant with regulations.
The improvements in outcomes for intermediaries and their clients using progressive data science solutions are considerable. Compliance budgets are increasing year on year for the majority of UK firms, but so are the volume and sophistication of criminal activities. Firms can make their resources go further still, by reducing the time taken to make decisions, whilst also making better customer verification decisions, and reducing the cost of data errors and duplications. Data science can even provide opportunities to allow previously excluded customers to gain risk-controlled access to products and services that are more likely to be suitable for the customer’s needs.
But, most importantly, the risk of individuals being able to commit fraud is significantly reduced. With the average cost of each fraudulent attack on small businesses amounting to £25,700 last year, this is not to be sniffed at. Cross-industry analysis of attributes enables unusual networks to be identified and disrupted, nipping the threat of financial crime well and truly in the bud.