Protect your customers with real-time transaction fraud detection

Arshi Singh, Product Director at ComplyAdvantage

Almost half of businesses globally have experienced fraud or a related economic crime in the last 24 months. What’s more, according to PwC, the risk of exposure grows as companies scale. 38% of businesses with less than $100m in revenue reported experiencing fraud, while the figure for firms with more than $10bn in revenue was 52%. 

Another compounding risk factor is innovation, with FinTechs such as neobanks and robo-advisers experiencing almost twice the rate of fraud compared to traditional banks. As newer entrants to the market, these firms are often less able to absorb the reputational damage that comes with negative news stories about defrauded customers.

Our State of Financial Crime report, based on a global survey of compliance professionals, also highlighted exceptionally high levels of concern about fraud during the pandemic. While the stimulus checks and relief payments that drove this have subsided, it remains a top predicate offense. 

Fraud detection and prevention is a vital tool in the arsenal of risk professionals looking to prevent fraud effectively. But in a dynamic environment where financial crime risks are increasingly bespoke to a firm’s business model and customer base, what does a best practice approach to transaction fraud look like? 

The challenge of fraud

Fraud analysts should consider their transaction fraud risk across three core areas of compliance:

Process: As new typologies emerge and criminal behavior changes, firms must align their enterprise risk assessments and identify any required changes to their business’ risk appetite. This may include evaluating new product lines, customer profiles, or regulatory requirements.

People: Fraud and anti-money laundering teams often work in silos. This can hinder effective financial crime risk detection. It may also leave patterns of illicit behavior that contextualize a potentially suspicious transaction undetected. Communication is critical, particularly as firms evaluate machine learning-based monitoring solutions. Data and information sharing across teams are essential to the success of these investments.

Platform: With an appropriate risk appetite set, and the correct internal alignment, fraud teams can take advantage of machine learning and behavioral analytics for fraud detection. In contrast to a rules-based approach, these solutions can project future risks and help teams anticipate threats. The straightforward configuration options offered by these tools mean it is possible to fine-tune alerts across various payment chains, responding to changing risks in near real-time.

Real-time fraud detection use cases

With the right approach across processes, people, and platforms, transaction fraud teams will be well placed to tackle fraud risks in their organization. ComplyAdvantage’s fraud detection solution underpins its Transaction Risk Management solution, providing real-time fraud detection across the key fraud typologies affecting financial institutions today.

In addition to rules that detect common fraud scenarios, we have an AI-driven approach that serves the risk and compliance team across a range of use cases. Here are some simple examples 

• A fraudster has a stolen card and wants to use it to withdraw cash. He uses the card in an out-of-pattern way, either withdrawing large amounts of money or using it in different countries. In this situation, the ComplyAdvantage solution will identify and flag the strange ATM transaction.
• A criminal buys a list of stolen cards online. Rather than attempt to use them for large purchases right away, she purchases small items first to ensure the card is still active and not flagged. Here, the ComplyAdvantage system detects the unusual pattern and flags the transactions.
• A customer typically only uses his account to buy clothes and groceries. Suddenly, there is a change in activity. A fraudster has used his card to buy designer shoes worth $500. In this scenario, the ComplyAdvantage solution detects the change in activity and flags the transaction.

Reducing fraud risk exposure: RealPage success story 

Fraud risk is nuanced and differs from industry to industry and even organization to organization. Working with ComplyAdvantage, financial institutions can target specific types of fraud through adjustable thresholds. Our fraud solution learns from the team’s actions and tunes the detection process over time to identify truly suspicious activity such as deviations in customer behavior or transactions that are physically impossible based on location. Our fraud solution will always explain why a transaction has been blocked or flagged for review so analysts can demonstrate the reasoning to regulators. 

For many firms, the data and risk scenarios their fraud prevention vendor uses are critical. Some solution providers have data sets heavily weighted towards one sector, such as banking. This can mean that potential risks in other industries are left out. That’s why RealPage, a leading global provider of property management software, chose to work with ComplyAdvantage. The flexibility to build custom scenarios helped the company implement unique rule sets needed to screen property managers and their tenants. The near-real-time results seen by RealPage enabled fraud alerts to be reviewed immediately and allowed team members to mitigate the risk of fraud losses for its clients.