Duplicate Pledging and Double-Factoring: How to Catch Invoice Fraud Before You Fund
By Zolvo Team · 8 min read
The two biggest credit stories of 2026 are both fraud stories. In the First Brands collapse, off-balance-sheet obligations hid leverage that lenders believed was around 5 times earnings but was closer to 20. In the Tricolor matter, the same collateral was allegedly pledged to more than one lender at once. Different mechanics, same lesson: the asset you think you are lending against may not be what, or where, you think it is.
For factoring companies and asset-based lenders, that lesson is not abstract. The invoice you fund today might already be financed somewhere else, might be inflated, or might not represent a real delivery at all. Fraud at origination, deciding whether an invoice is genuine before you advance against it, is now the single most important risk control in the business. This article covers the two fraud patterns that matter most, why factoring is structurally exposed to them, and what it takes to catch them before the money goes out.
What duplicate pledging actually is
Duplicate pledging, also called double-pledging, double-dipping, or double financing, is when a borrower finances the same receivable with more than one lender. The client sells an invoice to Factor A on Monday and the same invoice to Factor B on Tuesday. Both advance against it. Only one can be repaid from the single underlying payment, and the other is exposed the moment the scheme unwinds.
Its close cousin is the fabricated invoice: a document for goods never shipped or services never rendered, sometimes layered in among real invoices so it does not stand out. Industry fraud specialists consistently name these two, duplicate finance and fake invoices, as the most common attacks on receivables lenders.
In collateral-backed lending, the integrity of the eligible-asset base is the foundation of the loan. As one industry weekly put it after the year's frauds, the quality of the verification is the quality of the loan.
Why factoring is structurally exposed
Several markets have a central registry that makes double-pledging hard. Colombia has RADIAN. India routes much of its receivables finance through the GST network and TReDS. The United States has nothing equivalent. A US factor checks for competing claims through UCC filings, which are recorded state by state, are slow to surface, and are easy to miss when a borrower operates across jurisdictions or uses slightly different entity names.
The verification process has also lost its human signal. As debtors moved invoice approvals onto EDI portals, the direct factor-to-debtor phone call that once surfaced a fake or a dispute disappeared. Many factors now confirm only a small sample, spot-checking perhaps one invoice in twenty, because manual verification does not scale. That sampling gap is exactly where fraud lives.
Why the problem is getting worse
Generative AI has lowered the cost of a convincing fake to almost nothing. Fraud teams now report AI-generated invoices, fabricated proof-of-delivery photos, even synthetic photographs of buildings and facilities that do not exist, plus coached impersonators who answer verification calls. Catching a fake by eye no longer works when the document was generated by a machine to pass inspection. The volume and the quality of attacks are rising at once, which is why fraud has moved to the top of the agenda at industry conferences.
Where manual verification fails
Manual verification fails in three predictable ways. It samples instead of covering, so most invoices are never independently confirmed. It cannot see across your own book in real time, so the same invoice number funded twice in the same week slips through. And it has no durable record: when a loss happens and a funder or auditor asks what was verified and when, the answer is scattered across inboxes and call notes.
What modern fraud detection looks like
Catching these patterns before funding takes a few things working together.
- Verify before you fund, at full coverage. Every invoice, not a one-in-twenty sample, gets an independent confirmation with the debtor through the channel they actually use: email, portal, or an automated call.
- Detect duplicates across your own portfolio. The system flags the same invoice number, amount, and debtor appearing twice, across clients and across time, before the second advance goes out.
- Check the entity, not just the document. Domain age, business registration, and address consistency catch the two-month-old shell company and the facility that does not exist.
- Keep an evidence trail per invoice. Who was contacted, when, through what channel, and what they confirmed, recorded automatically, so the proof exists before you need it.
Verification is not a clerical step that happens after underwriting. It is underwriting. The factor that confirms every invoice, detects duplicates across its book, and keeps the evidence is the factor that does not become the next cautionary footnote.
Automate the coverage, keep the judgment
None of this means handing the decision to a model. The right design automates the coverage, the thousands of routine confirmations and duplicate checks no person can do at volume, and routes anything suspicious to a human with the evidence already assembled. Full coverage by machine, final judgment by a person, is both faster and safer than sampling by hand.
Zolvo automates invoice verification for commercial lenders, confirming invoices with debtors across email, portals, and voice, and flagging duplicates before you fund. See how invoice verification works, read why AI changes invoice verification, and review what counts as an ineligible receivable in a borrowing base. For the full operational picture, see how we support factoring operations.