4 technologies that improve fraud detection in banking

January 12, 2023

Banks and fintechs are facing fraud at a growing rate. From the pre-pandemic period in 2020 to 2022, there was a 42% increase in the average number of prevented and successful monthly fraudulent transactions at banks in the US.

To improve fraud detection in banking, financial service providers are leveraging technologies that use data to spot anomalies in customer behavior, fake documents, and other fraud schemes.

APIs that connect financial service providers to data source systems are a particularly powerful fraud detection technique. This is especially true for synthetic identities that combine real and fake information to avoid detection and are projected to cost organizations almost $5 billion by 2024. By accessing third-party customer data with an API from systems such as payroll, banks and fintechs can easily verify their customer is a real person and not a mixture of stolen and fabricated data.

With the help of technologies such as APIs and behavioral biometrics, financial institutions are now tackling some of the greatest challenges they’re facing in fraud today, such as the growing sophistication of fraud schemes.

1. Payroll connectivity APIs

With the customer’s permission, payroll connectivity APIs grant financial service providers access to the data inside their payroll platform. Since payroll platforms contain personally identifiable information and data on income and employment, financial service providers can use the API to tackle both identity and income fraud. Although this data is sensitive, Pinwheel research discovered that around 8 in 10 US workers feel comfortable sharing this data with financial providers.

Combating income fraud is critical considering its risk grew by 27.3% from Q2 2021 to Q2 2022. Even more concerning are the elaborate schemes scammers orchestrate to present fraudulent income statements as real. In some cases, they will have third parties pose as employers or provide months’ worth of bank statements with fake deposits.

By connecting to a customer’s payroll data via an API, financial service providers can quickly authenticate their identity and income and minimize the risk of fraud using data from source systems. Pinwheel Verify, for example, is an API solution that connects to over 1,700 payroll platforms to simplify identity, income, and employment verification.

Before implementing a payroll connectivity API, financial service providers should carefully consider the information security strategy and credentials of the API provider to avoid increasing another type of risk: data breaches. Pinwheel implements information security must-haves such as SOC 2 Type 2 and ISO 27001 certifications, bank-level encryption protocols like AES 256 and Transport Layer Security (TLS), round-the-clock systems monitoring, and a dedicated security and compliance team.

2. Artificial intelligence (AI)

Payment fraud is one of the most widespread fraud schemes in the financial industry, impacting 71% of organizations in 2021.

With AI technology, financial service providers can use the immense payment data they have at their disposal to build more effective solutions for fraud detection. These solutions reduce operational expenses and can learn how to detect new types of fraud, which allows them to continuously evolve.

AI is also able to detect false positives with more accuracy than rules-based fraud detection. Prior to implementing an AI-powered solution by Teradata, the Denmark-based Danske Bank only had a 40% fraud detection rate and as many as 1,200 false positives a day. With the AI solution in place, it reduced false positives by 60% and increased true positives by 50%.

3. Behavioral biometrics

While biometric data refers to fingerprints and facial images, behavioral biometrics uses more granular data for user authentication. Location, typing speed, and how a user holds their phone are behavioral biometric data points that can improve banking fraud detection in real time.

The speed of payment transactions has made it more challenging for financial service providers to detect fraud before it happens. Fraud investigation also requires resources – every $1 lost to fraud costs financial service providers $4.23. The solution lies in detecting fraudulent activity before the damage is done. Behavioral biometric solutions are able to analyze a customer’s behavior throughout the banking session and flag suspicious activity ahead of time, without damaging the customer experience.

A large US bank that implemented a behavioral biometrics solution found that familiar input behavior (the typical way a user logs in) was detected in 96% of trustworthy users, while a familiar typing pattern was seen in 91% of trustworthy users. Eighty-seven percent of users who were labeled as high risk had an anomalous input pattern, indicating that a script (rather than a human) was typing in the user’s credentials.

4. Social media profiling tools

Social media profiling tools can support fraud detection in banking by providing more data on the customer’s identity. By using publicly available information from social media, these tools help banks improve the detection of synthetic identities and check whether any of the customer’s information was exposed in a data breach.

A reverse email search, for example, returns data on different social media accounts associated with the email address. The same can be done with phone numbers. If an email or phone number isn’t linked to any social media presence, financial service providers can flag it as potentially fraudulent and review the customer’s profile more closely.

Banks and fintechs can also use social media data, such as the customer’s location or place of work, to cross-check the information they provided in their application.

Banking fraud is constantly evolving and demands a sophisticated strategy

Online banking has introduced a great deal of convenience, but it has also made it easier than ever before to execute phishing attacks and commit fraud and other financial crimes.

Criminals can buy stolen identity data on the dark web, while some offer fraud-as-a-service on encrypted messaging apps. Others teach would-be fraudsters how to avoid Know Your Customer (KYC) verifications and even provide lists of potential targets.

Although financial service providers cannot avoid being targeted by bad actors, they can ensure their fraud detection systems leverage secure technology that leaves little room for manipulation. Access to verified identity and income data from source systems, such as payroll, helps banks and fintechs save time on KYC processes and quickly determine whether their customer is who they say they are.

By using APIs to retrieve this information, financial service providers can still provide a seamless, digital user experience that has become the norm, without compromising security or increasing fraud risk.

With Pinwheel’s payroll connectivity API, you’re able to prevent fraud with verified payroll data. We keep your customers’ data safe with bank-level encryption protocols and a strong data security strategy led by a chief information security officer.

Our CRA status also prevents you from violating the FCRA while protecting your customers from incorrect data. Likewise, our team members have experience with highly regulated industries, including banking, and we have a bank deployment manager to support you through the API implementation.