57% of Firms Uncover Fraud After Settlement
- Most firms detect payment fraud after settlement, limiting their ability to intervene before funds move.
- Earlier verification and integrated workflows can improve payment integrity and reduce delayed-detection costs.
- OrboGraph applies layered detection to on-us and deposit fraud using transaction analysis, image forensics, rules, and analyst review.
Payment fraud is often measured by what was lost, but the more important question may be when the problem was identified. According to PYMNTS(opens in new tab), 57% of firms usually discover fraud or failed payments only after settlement, while just 30% identify those issues before initiation or during authorization.
That timing gap creates more than an operational inconvenience. The same PYMNTS Intelligence research found that payment fraud and nonclearance cost middle-market firms an average of 31 basis points of revenue, rising to 42 basis points among firms with higher uncertainty. In payment operations, a late alert can mean fewer options, more investigation, and a greater chance that funds have already moved beyond immediate control.
The lesson is straightforward: fraud prevention is most effective when it becomes an early-stage decision rather than a post-settlement investigation. For financial institutions, that means connecting transaction behavior, account intelligence, check images, and review workflows before a suspicious item is paid or funds become available.
The Cost of Detecting Fraud Too Late
PYMNTS reports that high-uncertainty firms experienced payment fraud and nonclearance costs twice as high as low-uncertainty firms—42 basis points of revenue compared with 21 basis points. Faster payments can intensify that exposure: firms that said payment speed increased their fraud risk reported accounts receivable integrity costs of 41 basis points, approximately 60% higher than other firms.
These findings point to a broader operating challenge. Fraud controls that sit outside the payment workflow—or that return an answer only after settlement—may provide useful evidence, but they do not provide the same opportunity to prevent loss. A control must be close enough to the transaction to influence the decision that matters: approve, hold, review, or return.
Earlier verification is one way organizations are closing that gap. Among firms that typically detect fraud or nonclearance before settlement, 81% use instant bank account verification, compared with 47% of firms that identify problems after settlement. The research also found a similar difference for open-banking ownership checks, at 76% versus 35%.
For check payments, the same principle applies. A financial institution needs more than a single rule or a basic image comparison. It needs layered analysis that can evaluate the account, transaction, image, and surrounding history quickly enough to support a decision before the item creates downstream exposure.
Building an Earlier Check Fraud Decision
Anywhere On-Us Fraud(opens in new tab) is designed for the paying bank, where inclearing checks can be evaluated for counterfeits, forgeries, and alterations. Its five-layer approach combines account profiles and thresholds, transaction analysis, image forensics, rules, and fraud review support so institutions can assess risk using more than one signal. On the deposit side, Anywhere Deposit Fraud(opens in new tab) extends layered detection across deposit channels. The module combines transactional and behavioral analysis with image forensics, consortium data, configurable rules, and fraud-review capabilities. It can operate as a scoring engine within an existing environment or run independently, with real-time and batch deployment options.
The objective is not simply to generate more alerts. It is to make each alert more useful by giving fraud teams context: why an item is unusual, which characteristics contributed to the score, and how the account or image compares with prior activity. That context can help reduce avoidable manual review while giving analysts better information for the cases that require judgment.
Closing the Timing Gap
The PYMNTS findings reinforce a practical fraud-prevention principle: the earlier an institution can identify risk, the more choices it has. A decision made before settlement can prevent funds from moving, reduce recovery pressure, and protect the customer experience from the disruption caused by a late fraud discovery.
For financial institutions evaluating their check fraud strategy, the next step is to examine where detection occurs today—and where it should occur instead. Anywhere On-Us Fraud(opens in new tab) helps address risk at the paying bank, while Anywhere Deposit Fraud(opens in new tab) supports detection across deposit channels.
Payment fraud will continue to evolve, but institutions do not have to accept after-the-fact detection as the default. By combining earlier decisioning with layered check intelligence, financial institutions can move from investigating completed losses to preventing more of them before settlement.