Can this tell me whether a pay stub is fake?

No tool can prove that from the file alone, and this one does not claim to. What it can establish is what the file's own structure records: whether a revision was appended after the payroll system wrote it, whether an amount sits in a different font than its neighbours, whether a consumer editor touched a document that claims to come from a payroll platform, and whether the metadata dates agree with each other. Each finding ships with its evidence and its innocent explanation, and the decision stays with you.

What does an edited pay stub look like inside the PDF?

The commonest edit is an appended revision: the payroll system wrote the file once, and the editor saved it again by adding a new section to the end, leaving the original bytes in place. The file then carries two versions of itself, and the earlier one is recoverable. Other traces include a second font subset for a typeface where one amount was retyped, an editor's name where a payroll system's should be in the producer metadata, and creation and modification stamps that disagree.

What kind of tampering does a byte-level check miss?

A stub generated from scratch (printed to a fresh PDF, rebuilt in a template, or produced by a pay stub generator), carries none of the editing traces this page reads, because nothing was edited: the file is cleanly written, just not by an employer's payroll system. Catching that class needs a reference for what the genuine issuer's output looks like, or verification with the employer. This page says so rather than implying coverage it does not have.

Is the pay stub I upload stored anywhere?

No. The file is sent over HTTPS, parsed in memory and never written to disk. The report is returned in the response and the document is discarded with it.

Pay Stub Checker: Was It Edited?

Drop a pay stub here, or browse

PDF or photo · Max 10 MB · processed in memory, never stored

No pay stub handy? Try the samples:

Three fictional documents: a utility statement, because a real employer's pay stub cannot be published, and the mechanics are identical: the edited copy has a second revision that changed the printed total, and the signed copy was altered after signing.

What this checks

Most edits to a pay stub: a raised gross, a changed employer, a shifted pay date. Are made by opening the payroll system's PDF in an editor and saving it. That leaves traces the editor cannot help writing:

A PDF grows by appending, so an edited pay stub still contains the page payroll wrote.

This is the shape of an edited file, not a finding of fraud. An HR system that stamps, merges or watermarks a payslip appends a revision the same way an edit does, and the report names the benign cause next to every signal.

An appended revision. The payroll system wrote the file once; PDF editors save by adding to the end of it, not rewriting it. The original page, the real amount, is still in the bytes, and the report says what changed between revisions.
The wrong toolchain. Payroll platforms print through server-side libraries. A stub whose metadata names a consumer PDF editor is a fact worth explaining, and the report separates tools that create documents from tools that consume one.
An amount in the wrong font. Retyping one number embeds a second subset of the typeface, a signature of text added with a different tool. On single-revision files this is reported as information rather than escalated, because ordinary generators can produce it too, that recalibration is published, with the false-positive study that forced it.
Metadata that argues with itself: a creation stamp from one day and a modification stamp from another, ID pairs and toolchain claims that do not agree with the file's own revision count.

What a signal does, and does not, mean

Every finding is reported with its evidence and its benign explanation, because every one of these traces has an honest cause too. An appended revision is also what an HR system that post-processes its PDFs produces: stamping, merging or watermarking a payslip appends a revision exactly the way an editor does. A stub that went through a scanner has a scanner's structure, not a payroll system's. A stub printed to PDF by the employee, because that was the only way to save it, is a cleanly written file that says nothing either way. This page reports signals, never a fake/genuine verdict. The reader who knows where the document came from is the only one who can weigh them.

The limits run the other direction too. A stub generated from scratch (a template rebuild, a pay stub generator) carries no editing traces, because nothing was edited: the file is cleanly written, just not by an employer. And a photographed stub has no revision chain or fonts to read; the image engine reports what image metadata and pixels expose, which is less, and the report says which engine ran. A quiet result is evidence, not proof.

For landlords, HR and lenders

Edited pay stubs turn up in income verification: a rental application, a background check, a loan file. Signals mean questions, not a caught forger: ask for the original download from the payroll portal, or verify employment at the source. A quiet result is consistent with an untouched file, and is also what a competent re-generation looks like. For a decision that matters, the employer's confirmation outranks any file inspection. The field guide walks every signal this page can raise, with real examples.