Document Fraud Detection Software: How Automated Checks Work — and Their Limits
What document fraud detection software actually checks — metadata, structure, pixels, visual layout, logic and history — how to evaluate a vendor, and what no tool can do.
If your business makes decisions based on documents — leases, loans, refunds, claims, payouts — you have probably considered automating the review. Here is what these tools do, and what to ask before choosing one.
The layers of a good check
- Metadata — which software created and changed the file.
- File structure — incremental saves, overlays, embedded fonts, digital signatures.
- Pixels — compression differences that reveal edited regions.
- Visual analysis — fonts, alignment, layout and the issuer's known design.
- Logic — arithmetic, dates, check digits, tax rates and running balances.
- History — whether the same file was submitted before.
What to ask a vendor
- Explainability — does every flag point to evidence you can see?
- Location — are findings marked on the document?
- Coverage — which document types and issuers are supported?
- Privacy and security — encryption, retention, where data is stored, and whether your documents train models (and how to opt out).
- Integration — dashboard and API, with structured output.
- Honesty about limits — accuracy claims you can test on your own documents.
What no tool can do
No detection system is perfect: there will be false positives and missed fakes. A tool cannot confirm that money moved or that an employer exists — only the issuer can. Use automated checks to decide what to verify, keep a person in the loop, and apply the same process to everyone.
How ThisVerify approaches it
ThisVerify runs all six layers, marks every finding on the document with an explanation, and gives a report in under a minute — with plans from ₪39 (about US$10) a month.