Using analytics to alleviate fraud risks in financial services

Attributed to Adam Mayer, Senior Manager, Technical Product Marketing, Qlik | 8 February 2021

Adam Mayer

Senior Manager, Technical Product Marketing, Qlik

Fraud typically involves deceit with the intention to gain at the expense of another illegally or unethically. Everyone can be vulnerable, as bad actors target both consumers and organisations alike. There are many types of frauds that consumers and organisations in the financial sector can be exposed to. For example, consumers can suffer from credit card frauds where someone steals their card details and misuses it, while banks can fall for loan frauds when customers borrow money from them with no intention of repaying the debt. Insurers aren’t spared as well and can be victims of insurance claim frauds where customers make claims for theft or breakage of items which have not actually been stolen or broken. 

Fraud is becoming more prevalent in this increasingly digital world. With a focus on digital payments, financial criminals have also shifted their targets online, as seen in the recent spate of phishing scams involving OCBC Bank in Singapore where 790 customers lost a total of S$13.7 million in one month. Globally, the number of online card fraud attempts increased by 23%, according to Feedzai’s Financial Crime Report Q3 2021 Edition. Therefore, it is important for financial services to stay on top of the potential risks they or their customers may be vulnerable to. Organisations must put in place processes and adopt the set of right tools to monitor and predict any potential risks for them to act quickly and nip it in the bud. Here is where data and analytics tools can help.

Analytical products can help organisations learn about the trends based on data around the types of frauds that are occurring, its frequency and severity to enable them to patch the potential gaps in their products or services. When it comes to fraud analysis, traditionally, financial institutions would have a set of rules in place that would examine requests and offer a decision to proceed with the request (or not). Unfortunately, as more rules are constantly added, these rules-based anti-fraud systems become very complex, and they don’t always adapt to hidden threats. This sometimes results in too many false-positives – blocking legitimate transactions while missing out on fraudulent transactions. 

On the other hand, analytics together with machine learning (ML) provides the ability to collect massive amounts of disparate data, analyse that data at scale and in context, and assign a risk score in real-time. This enables a risk-based fraud analytics solution to apply the precise level of security, at the right time, through step-up authentication. At Qlik, we have been equipping our customers with analytical tools and solutions to test and assess the adequacy and effectiveness of the business and IT controls that are in place to mitigate fraud. Examples include tests to identify unusual transactions, assess the frequency and value of transactions such as those that are above or below specific limits, and consumer behaviours that are out of the norm.

Implementing data and analytical tools to aid them in fraud prevention can be complex to some, especially small financial businesses. Here are some best practices on fraud analytics that can help organisations navigate through the process.

  • Use all available data to conduct fraud tests and ensure that these data are VACANT (Valid, Accurate, Complete and Nicely Timed).
  • Ensure that the analysis of the data achieves the desired purpose and use care in aggregations as it is easy to make mistakes in formulas. History is littered with cases where analysts made minor errors in formulas that caused major errors in results.
  • Instead of showing analysis results in tables, use visualisations to help tell a story with the results to create impact and drive change.

By leveraging data and analytics, organisations in the financial sector can take on a proactive approach to mitigating fraud based on the trends and predictions. Beyond enhancing the backend systems, organisations should also continuously educate consumers about prevalent scams, how to avoid them, and what to do if they become victims. Banks specifically, should create an open line of communication for consumers to escalate any suspected fraud cases so it can be dealt with in a timely manner.

The effectiveness of fraud analytics, along with consistent engagement with consumers, can potentially save organisations millions from fraudulent incidents and allow them to focus on the things that matter most – strengthening the trust between its customers and the brand.