How False-Positives Complicate The B2B Payments Fraud Fight

Retailers focused on combating fraud have credit cards in the cross-hairs of their efforts. Between data breaches exposing customer details and card information and the rise of card-not-present fraud as operations move online, digital businesses are challenged to stay abreast of payment security trends — and fraud is a massive issue for firms large and small.

But ramping up the war on card fraud can introduce a new risk to companies: false positives. According to PYMNTS data, more than 60 percent of digital platforms say too many false positives are a significant point of friction in the conversion process — and more than 30 percent say it’s their number-one challenge.

According to the latest Payments 2022 Playbook: Building A High-Performing Payments Team For Fraud Detection, a PYMNTS study in collaboration with Stripe, digital platforms continue to express discontent with their current fraud strategies, and false positives are compromising their brands, customer relationships, and bottom lines.

This isn’t merely an issue for the B2C world, however. Indeed, as B2B organizations begin to pay closer attention to the rising threat of fraud in the procure-to-pay process, the issue of false positives can result in similarly negative consequences, from declined card payments to broken vendor relationships.

The False Positive Threat

In accounts payable and accounts receivable, rising fraud mitigation efforts may also lead to more false positives, leading to declined commercial card transactions, misplaced suspicion between buyers and suppliers, delays in invoice processing and more. And just as false positives can jeopardize the business-consumer relationship, so can they negatively affect the buyer-supplier relationship in the accounts payable and accounts receivable context.

In a research report published in the December 2018 International Journal of Accounting Information Systems, authors Galina Baader and Helmut Krcmar examined some of the strategies that organizations can deploy for reducing the instance of false positives in fraud detection, with a particular focus on B2B processes like procure-to-pay and accounts payable. Among the largest threats in this area is redirect payment fraud, in which a bad actor infiltrates company emails or systems to change bank account details, payment amounts, and other key information to redirect B2B payments to a different account for personal gain.

Cybercriminals may slightly change the name of a legitimate vendor to trick a company into paying a seemingly real company, or a vendor can suddenly change currency between purchase and payment to take advantage of a payment conversion. Suppliers can also submit duplicate invoices, or issue the same invoice with different billing amounts.

These accounts payable-related warning signs are among the many so-called “red flags” that fraud examiners watch for when looking to detect nefarious behavior, according to Baader and Krcmar. According to their research, by combining the red flag approach with the processing mining, which extracts data from actual processes and their event logs, the instance of false positives can be significantly reduced.

Mixing Humans With Technology

This strategy, however, requires a deep understanding of, and appropriate technology adoption for, data analytics capabilities within the enterprise.

In a Q&A session last year with SAP Insider, Ingo Czok, CEO of Tembit Software, which was acquired by Hanse Orga Group, explained that a combination of human interference and machine learning can be an effective strategy in not only identifying fraud, but reducing the number of false positives.

“It is always difficult to fine-tune your rules to match each and every real fraud and not come up with too many false positives,” Czok explained. “[Machine learning] will help with generating new rules based on previous fraud or non-fraud decisions. Therefore, it is best practice to have a fraud monitor to actually view the transactions with a high fraud score and manually accept or deny those to further train the AI through that process.”

“This is precisely the reason it is important to have human engagement in the process,” he added, noting that the balance of automated data analytics with human expertise will result in greater overall fraud prevention, as well as the ability to mitigate false positives.

The remarks offer additional evidence that supports digital organizations’ investment in expanding their teams to promote fraud detection and avoidance of false positives. As the Payments 2022 Playbook examines, digital platforms should also turn to their third-party vendors to foster an environment of collaboration in the fight against fraud — which, PYMNTS research suggests, may also combat false positives.

The issue of false positives is quickly becoming the biggest fraud concern for digital platforms. With the B2B ecosystem getting up-to-speed with the threat of B2B payments fraud, security initiatives will have to balance their fraud mitigation efforts with the ability to avoid false positives to keep business-to-business transactions running smoothly — and to keep the buyer-supplier relationship intact.