At the start of a new year, we often see an increase in invoice finance fraud – when clients seek to borrow money by supplying lenders with inaccurate information. (Not to be confused with invoice fraud which sees fraudsters send false, real-looking invoices to extract payment from companies.)
The mechanics of fraud differ for invoice factoring and invoice discounting. But at the heart of it sit false, duplicated or re-aged invoices, fake credit notes, diverted receipts and ledger reconciliation challenges.
Invoice finance fraud becomes harder to spot during peak sales periods. Because the rise in trading legitimises an increase in borrowing. And as lenders’ resources get stretched due to high volumes of borrowing requests – fraud can slip through.
To spot and stop invoice finance fraud promptly lenders need to have the right tools. This can be done by using intelligent automation across the entire process, from lending through to account management and controls.
1. Improving the review processes
For invoice factoring, a manual review process means a level of human error is inevitable. The result is red flags get missed while hefty due diligence processes become more time-consuming.
Automating exception management can improve the review process. This sees automations (or bots) run background checks on an invoice to spot irregularities across invoice number, size or assigned credit notes. If something unusual is detected the relevant teams are notified and asked to investigate further. Here human action only happens when necessary, saving employee time while reducing the margin for error.
2. Enhancing visibility
Another challenge for invoice factoring companies comes from diverted receipts – when payments collected by the client aren’t transferred to the lender. This can go missed until the client’s debt extends the agreed terms or an aged, outstanding invoice gets flagged.
Automation can be used to recognise this fraud quickly. Bots can monitor the client’s banking history for anomalies. For example, a bot can single out when payments from named debtors aren’t passed back to the lender on schedule.
3. Increased verification
Verifying invoices, new borrowers and their creditors is crucial for invoice factoring and discounting. A simple step such as checking that the goods or services have been received can become a bottle neck when manually chasing proofs of delivery.
Automation makes verification quick and easy. Here bots can automatically send emails to request documents, chase customers and process the data shared by extracting relevant information. They can flag discrepancies and unusual activity for the team to review. So, people don’t need to do low-value activities such as chasing and can focus on tasks that require detailed reviews.
Bots can also verify new borrowers with more speed and accuracy. Same directors being named on borrower and creditor boards is a tell-tale sign of fraud. Bots can scan online data sources to spot any correlation between borrower and creditor directors. This is smart way to stop fraudulent applications in their tracks.
4. Fast-tracking the ledger reconciliation process
Efficient ledger reconciliation is essential for invoice discounting lenders – sadly it is often slow and error prone. First, it takes clients two weeks to submit their ledgers, then it takes lenders ten more days to manually reconcile these with their records. Meaning any mismatch and potential fraud are detected almost a month later.
Automation can help with reconciliation in two ways.
First, a workflow can be set up to pull clients’ ledger automatically at the start of each month. Second, bots can review entries across clients’ and lenders’ ledgers to detect discrepancies and animalities. These are then handed to people to investigate further. This ensures fraud is spotted earlier in the month, and action is taken straight away.
5. Harnessing advanced technology
Automation is a great stepping-stone to doing more with data – it improves data quality. This data can then be used to get in-depth insights by deploying analytics or machine-learning.
For example, analytics can unearth more patterns that signal suspicious activity. While machine learning can look at external and contextual data sources to determine those tricky fraudulent transactions that go undetected due their more elusive nature.
The best thing about automation is that it sits on top of lenders’ existing IT infrastructure with minimal disruption to existing systems. There is no need to rip and replace applications or to learn how to use a new process. Bots are calibrated to work with lenders’ systems and processes, requiring minimal IT involvement.
Automation is not just a more efficient but also a non-intrusive way to spot and stop fraudulent activity. With this this solution in place, lenders can keep fraud at bay even during the busiest periods without straining resources of increasing spend.
This article is written by Venugopala Dumpala, Practice Head Banking & Financial Services at Firstsource in Global Banking and Finance Review.