Document fraud detection

Detect a fake proof of address

Utility bills, phone bills or rent receipts: proof of address is required to open an account, rent a home or complete an administrative procedure. Easy to retouch, it is frequently forged. Docusan analyzes the document you receive and flags modifications.

Why is proof of address so often forged?

Proof of address opens many doors: a bank account, a home, a loan, an identity document or a contract. It is also one of the easiest documents to retouch, as it is usually an ordinary bill received as a PDF: changing an address or a name takes no particular skill.

For whoever checks it, the difficulty is that there is no single format: utility bills, phone bills, rent receipts and insurance certificates all look different, and both the accepted documents and their maximum age change from one procedure to another. A serious check therefore follows a method rather than a quick look.

How to detect a fake proof of address: a 6-step method

  1. Compare the name and address with the rest of the case file

    The account holder must match the person, and the address must match the other documents: ID, tax notice, employment contract. A different spelling, house number or postal code deserves a question.

  2. Check the date and how recent it must be

    The limit depends on the procedure. In France, for example, it is less than a year for an ID card or passport and less than 3 months for registering on the electoral roll. A recent date with an old billing period should raise a flag.

  3. Read the 2D-Doc code if there is one

    On French documents, this two-dimensional barcode holds the document’s data and an electronic signature. A certified reader lets you compare what is signed with what is printed. Our 2D-Doc guide explains how.

  4. Check the issuer

    Does the supplier exist, and are its details consistent: company name, registration number, address, customer number? In France, look the company up in the Annuaire des entreprises. To contact the issuer, use its official contact details, never those printed on the document.

  5. Examine the file, not just how it looks

    A PDF keeps traces: the software used, creation and modification dates, saved versions. A bill from a major supplier that comes out of an image editor is questionable. Our free tool reads these properties in your browser, without sending anything, and our guide explains how to tell if a PDF was edited.

  6. Cross-check before concluding

    A single clue proves nothing. What counts is how many there are and whether they fit the rest of the case file: when in doubt, ask for another document rather than accusing anyone.

In doubt about a document?

Analyze it for free: Docusan flags the anomalies and locates them on the document.

Try it for free

Warning signs to look for

None of these signs is proof on its own; several together justify a closer check:

  • Layout: a different font, size or alignment on the address, name or date compared with the rest of the document.
  • Quality: an area sharper or blurrier than the rest, a slightly different background around a piece of text.
  • Consistency: a billing period that does not match the date, a postal code that does not match the town, amounts that do not add up.
  • Issuer: an approximate logo, missing legal notices, an unusual customer service number.
  • File: a screenshot or photo where you would expect the original PDF, for no apparent reason.

Rent receipts and people staying with a host: special cases

A rent receipt is the easiest document to write from scratch, and one made on a word processor is nothing unusual: many landlords work that way. Check instead that the landlord, the home and the lease are consistent. Depending on the procedure, only some receipts are accepted: in France, a driver’s license application does not accept handwritten ones.

A person staying with a host has no bill in their own name. For a French ID card or passport, they provide a signed statement from their host, a copy of the host’s ID and a proof of address in the host’s name. In that case, it is the host’s document that needs checking.

A genuine document can also be wrongly refused: for French identity documents, service-public.gouv.fr states that a secure proof of address bearing a 2D-Doc code cannot be refused.

What to do if the document looks forged

Start by asking for an explanation or another document: a technical anomaly can have a legitimate cause, such as a re-saved file or an error by the issuer. Keep a record of your check and only decide on a case file on the basis of all the evidence.

If forgery is established, legal consequences may follow. In France, article 441-1 of the Criminal Code punishes forgery and the use of forged documents with three years in prison and a €45,000 fine. What to do next, such as filing a complaint, is for your legal department. Our guide on document fraud says more.

Automating the check as volumes grow

These checks take time, and the eye does not see everything: a careful edit can leave no visible trace. Docusan analyzes the file you receive, as a PDF, scan or photo, and gives a verdict: No detection, Warning or Alert. The anomalies are located on the document, with a comment to understand them, and the decision always stays with your teams.

Docusan does not contact the issuer or query government databases: it complements the checks above and does not replace them. You can use it from the platform or integrate it into your tools with its API, with the Enterprise plan, and your documents are never used to train our models (security and privacy).

Sources

Typical frauds

The most common forgeries

01

Edited address or name

The address or the account holder’s name is replaced on a genuine bill to prove a residence that is not theirs.

02

Fabricated rent receipt

A rent receipt is written from scratch for a home or a landlord that does not exist.

03

Re-dated bill

The date of an old bill is changed so that it looks recent and meets the required time frame.

Our detection approach

What Docusan analyzes

Four levels of analysis to spot forgeries, whether the document is a PDF, a scan or a photo.

01

Metadata

  • Missing metadata
  • Inconsistencies in the file properties
02

Pixel-level anomalies

  • AI-generated images, deepfakes
  • Inserted content (splicing, inpainting, copy-move…)
03

Structural anomalies

  • Edits and annotations
  • Recovery of previous versions
04

Customizable detectors

  • Consistency with document templates
  • Data cross-checking
  • Watchlists

The result

What a report looks like

Each document gets a verdict: No detection, Warning or Alert. The anomalies are located on the document, with a comment to understand them.

Alert

  • Modified area of the document
  • Inconsistent file properties
  • A comment for each clue
Illustration with a fictitious document. The clues depend on the document analyzed.

FAQ

Your questions

How can you tell if a proof of address is fake?

Compare the name and address with the rest of the case file, check the date, read the 2D-Doc code if there is one, check the issuer and examine the file itself: software used, dates, saved versions. A single clue is not enough: what counts is how many there are.

How recent must a proof of address be?

It depends on the procedure and the organization asking. In France, it ranges from less than 3 months for the electoral roll to less than a year for an ID card or passport. Outside official procedures, each organization sets its own rule.

Is a screenshot of a bill suspicious?

Not necessarily, but it deprives the check of the original file’s information. If in doubt, ask for the PDF downloaded from the issuer’s customer area.

Which proofs of address can Docusan analyze?

Any proof received as PDF, scan or photo: utility or phone bills, rent receipts, certificates. The checks depend on the file format.

Is a rent receipt written with a word processor necessarily fake?

No: many landlords write their receipts themselves. Docusan flags technical anomalies in the file, and your teams interpret them in context.

How is the result presented?

Each document gets a verdict (no detection, warning or alert) and the anomalies are located on the document in the viewer.

Try it for free

High volumes?

Automate your checks

A subscription with included pages from the Pro plan; the API, dedicated infrastructure and customized detectors with the Enterprise plan.