Case Study: Ultimate Finance
Instant bank statement checks on every deal, so altered documents get caught early.
We’re trusted by the world's leading companies
of fraudulent lending stopped across a small sample of cases
saved per deal, allowing valuable time to be spent elsewhere in the review
Up to 10% of documents contain risk signals
Fraudfinder is a valuable tool for us and we use it on every deal. It helps us review applications more quickly, especially where we don’t have CATO data, and supports a more efficient, confident decision-making process.
Protecting Ultimate Finance’s loan book against document fraud
Fraudfinder helps Ultimate Finance verify business bank statements in under three seconds on every funding application, saving time and catching falsified documents before they reach the loan book.
Ultimate Finance is a specialist asset-based lender that has been supporting the ambitions of UK businesses since 2002, providing over £19bn in funding to date. Backed by the Tavistock Group, it offers a flexible range of funding solutions including Invoice Finance, Asset Finance, Bridging Finance, and Structured Finance.
Ultimate Finance is solution-led: business needs come first, with funding tailored to each client. That same care extends to its underwriting, where confirming the authenticity of customer bank statements is critical to protecting both the business and its loan book.
As a business finance provider, Ultimate Finance’s underwriting team receives bank statements from virtually every banking provider in the UK, including: Tide, Starling, Revolut, Zempler, Mettle, Counting Up, ClearBank, Anna and all high street banks.
With fraud cases still on the rise, and new technologies making it easier to alter statements, the team needed a tool to support underwriters in quickly and accurately reviewing statements provided for funding applications and spot any potential fraud early.
Automated document fraud detection on every application
Ultimate Finance uses the Fraudfinder portal to run instant, consistent checks on business bank statements submitted with funding applications. Fraudfinder analyses each document against several visual and structural fraud signals, and an adaptive database of genuine institution templates, returning a clear risk decision in less than four seconds. If a result needs clarification beyond automation, Ultimate Finance is supported by an expert human-in-the-loop review as a second opinion.
Fraudfinder is now used for every deal. Where statements pass in full, the team no longer needs to manually compare them against internal guides, and where a statement is clearly altered, Fraudfinder adds data-backed weight to confirm the decision and issues a document fraud report.
Example 1: Altered figures on a HSBC bank statement, with the originals restored
Our first example is a statement where the numbers had been changed. What makes it useful is that Fraudfinder doesn't just flag that something looks off; it can restore the data that was deleted or written over, giving you a side-by-side of the original file and the version that was submitted.
The text in blue is data that was in the original document but removed before it was sent. The text in orange was added after the file was first generated. Together, they show exactly what was changed and where; in this case, the transaction figures and balances.
Fraudfinder provides data points the human eye can’t see, so the underwriter can make the call with the evidence in front of them.
In this example, the applicant changed the address on their statement. What they didn't realise is that many bank statements carry a barcode that encodes the customer's real details.
Fraudfinder's barcode decoder read the barcode on the document and pulled out an E13 (London) address. The address printed on the statement showed a BS1 (Bristol) postcode. The two didn’t match, which tells us the visible address was changed after the statement was issued, while the barcode still held the original.
It's a good reminder that the parts of a document a fraudster can see and edit are only half the story. Fraudfinder reads the parts they can't.
On this Allica Bank statement, Fraudfinder detected 93 separate edits to the file!
Almost nothing on the page was left untouched. The opening balance, the dates, the transaction descriptions, the running balances and the address had all been changed. At that point, you're not looking at a genuine statement with a couple of tweaks; you're looking at a document that has been rebuilt.
A statement like this would take real time to pull apart by hand, and all edits may not be as easy to spot. Fraudfinder surfaced all 93 automatically and flagged the document as high risk.
About
Fraud & risk in mortgages and account ownership
Fraud & risk in business lending
Features applicable:
Advanced document fraud detection
Bank statement verification
Confirmation of Payee & account ownership
Metadata & tamper analysis
Font & structural analysis
See Fraud Finder in action
Instant fraud detection with the click of a button...
Looking to automate your fraud risk process?
Book a demo today...
We'll provide a short no-obligation demo of each element of our fraud tech, book below:
