Escalating payments fraud requires advanced treasury platforms

Respondents reported an average profit loss of 1% from fraud and compliance violations in 2021. And the larger the company, the more criminals will target it. According to the Association for Financial Professionals (AFP) 92% of finance leaders observed that fraud in 2021 was as bad as, if not worse, than the year before. The problem appears to be worsening: A recent PYMNTS survey found 59% of financial institutions (FIs) experienced an increase in their overall fraud rate compared to the previous year. CFOs and treasurers clearly require a more complete set of payment controls to stop fraud altogether.

Additionally, it’s getting harder and harder to protect against fraud. One lingering effect of fraud during COVID is that the pandemic introduced a whole new generation of fraudsters. This generation has access to inexpensive low- and no-code technology that makes it extremely easy for them to carry out their tactics.

Hybrid and remote work might also be helping accelerate fraud trends. According to a survey from the Canadian cybersecurity firm CIRA, 55% of cybersecurity professionals view their organisations as more vulnerable to cyberattacks because some, if not all, of their employees work remotely. The heightened risk stems from remote workers often using personal devices, insecure Wi-Fi networks and file-sharing platforms that are less secure than their corporate counterparts.

For example, business email compromise (BEC) scams continue to plague treasury and finance departments. Wire fraud also remains a significant problem as it is most susceptible to fraud. Another area is in ransomware attacks where a company’s internal system is compromised (usually through phishing) and taken over. Users are instructed to either pay a ransom or permanently lose access to their systems.

To combat the threats of today, fraud prevention solutions should include these capabilities:

• Automated payment processes to standardise controls
• Real-time screening of all payments data to identify suspicious transactions
• User-defined payment screening rules
• Resolution workflow to investigate suspicious payments
• Monitoring of the status and priority of alerts in a KPI dashboard

Tools like artificial intelligence and machine learning (AI/ML) as well as application programming interfaces (APIs) are game changers in the fight against fraud. At Digiata we understand and use these technologies. Modern payments fraud detection software like Kyriba’s Payment Fraud Detection module offer these solutions and more. Its advanced tech allows you access to the following:

Real-Time Screening, Alerts and Notifications

The rise of same-day and real-time payment systems has increased the need for real-time responses to fraud attempts. Modern fraud detection software uses artificial intelligence (AI) and machine learning to screen payments against historical payment data, pinpointing any anomalies. By providing more complete data, these solutions enable data-driven decision-making.

Isolation Forest

An isolation forest model is an unsupervised algorithm that works on the principle of isolation anomalies; anomalous instances in a dataset tend to be easier to separate from the rest of the sample.

Generative Adversarial Network

One problem many machine learning models run into when trying to identify fraud is the lack of data around fraud. Most organisations haven’t experienced significant payments fraud and thus lack the depth of examples to share. Generative adversarial networks (GANs) can solve this problem. A GAN is a deep learning model that pits two separate neural networks against each other. One network (the generator) mixes real data and synthetic data together and attempts to outwit the opposing network (the discriminator).

Dashboards

Dashboards can be set up to display all suspicious payments and prioritise their resolution, based on factors such as detection rules, risk exposure, incident counts and a fraud detection scorecard. Dashboards provide authorised users with complete transparency into all payment screening and can resolve outstanding actions efficiently.

Fraud Prevention Workflows

Modern payments fraud modules also support fully automated, end-to-end workflows for the resolution of outstanding suspicious payments. Users can also determine how each detected payment should be managed, enforcing the separation of duties between the initiator, approver and reviewer of a detected payment.

Reporting and Audit Trails

Leading technology solutions can ensure that detected payments are permanently tracked in the system for daily, monthly or annual reporting. History is maintained indefinitely and all details of the suspicious transaction—including the audit trail of detected and resolved actions—are retained for internal and external audit reporting.

Payment Hub

Organisations have all their fraud protection capabilities in one place. Payment hubs consolidate payment streams from ERPs, finance, treasury, legal, capital markets and decentralised teams, transforming disaggregated processes into a single source of record for all outgoing payments. A payment hub also transforms payment data into bank-specific file formats and connects directly with global banks via multiple protocols, including host-to-host, SWIFT and regional networks. Payment Hubs Payments from ERPs or other systems can bring an entire enterprise-wide payments landscape under the consistent and risk-focused fraud prevention framework. With API-driven connections and integration tied into an approval and fraud prevention workflow, controls and fraud are enhanced and easily governed.

APIs

These bring unprecedented visibility and transparency to both the payer and the payee. However, there is also no stopping the transfer of funds once a real-time transaction is executed. Therefore, fraud needs to be prevented in the approval process before a payment request reaches the bank. Building APIs into the payment platform allows users to fully automate bank account validation and payment policy screening, identifying exceptions in real-time. APIs can instantly match payments against third-party data

If there was ever a time to invest in making your payments processing more secure, it’s now.

‍