How to Sign a PDF Online for Free

When you draw, type or upload a signature here, the result is an image or a line of text placed onto the page and written into the document's content. It is not metadata, it is not a certificate, and it is not attached to your identity in any way a computer can check. It is a picture of a signature, in the same sense that a scanned paper signature is a picture of one.

That is worth saying plainly because the word "signature" carries assumptions from the cryptographic world. A visual signature does not prove who applied it, does not prove when, and does not reveal whether the document changed afterwards. Anyone with the file can crop the mark out or place it on a different page.

A digital signature in the technical sense is a cryptographic operation. It binds a document to a certificate issued to a specific signer, and it produces a value a verifier can check to confirm two things: which certificate signed it, and whether a single byte has changed since. Alter the document afterwards and the check fails loudly.

Producing one requires a certificate and a signing key, which is precisely what a browser tool with no account cannot give you. So this workflow does not offer a digital signature, and no setting in it turns a drawn mark into one. If a counterparty asks for a digitally signed or certificate-backed document, a visual signature will not satisfy that request no matter how it looks.

There are three ways to produce the mark. You can draw one with a pointer or a fingertip, which is captured as a transparent PNG and supports undoing individual strokes while you get it right. You can type one, which is rendered in a standard font. Or you can upload an existing image of your signature, as PNG or JPG.

The choice affects how the signature looks, not what it is. All three produce the same kind of artifact and carry the same weight. Drawing on a touchscreen usually looks the most natural; an uploaded scan is the most consistent if you sign a lot of documents; typing is the fastest and looks the least like a signature, which for some recipients is a feature rather than a flaw.

This asymmetry catches people out, and it matters more than the signature question for anyone filling a form. The signature and any free-placed text you add are drawn into the page's content, so they are baked in: a recipient cannot click them, edit them or delete them without editing the document itself.

Values you type into a PDF's existing form fields behave completely differently. Those remain live, interactive form fields in the saved file, which means the recipient can click into them and change what you entered. Your signature stays put while the amount above it quietly does not.

That is what the flatten option is for. Flattening converts the filled fields into ordinary page content, so the values become as fixed as the signature is. Leave it off when the document genuinely needs to stay fillable — you are sending a partially completed form onward, for instance — and turn it on whenever the filled values are final.

The honest answer is that it depends on who is receiving the document and what they will do with it, and that is a question for them rather than for a tool page. What can be said usefully is which direction the answer usually runs.

A visual signature is generally fine where both parties already know each other, where the document records agreement rather than creating a legal instrument, and where nobody will later need to prove the signature's origin to a third party. Internal approvals, delivery acknowledgements, school and club forms, and routine supplier paperwork usually sit here.

It is the wrong instrument where identity has to be provable to someone who was not in the room. Anything involving property, formal witnessing, regulated financial or medical consent, or a counterparty who has specified a signing platform belongs with a dedicated e-signature service or a certificate-based signature. Those services exist to produce an audit trail — timestamps, verified identity, tamper-evidence — and that trail, not the appearance of the signature, is what they are actually selling.

If the PDF already carries a digital signature from someone else, adding anything to it — a signature, a filled field, a date — modifies the document and invalidates that signature. This is the system working as intended: the whole point of the cryptographic check is that it fails when bytes change.

So do not countersign a certificate-signed document by editing it here. Ask the sender how they want the counter-signature applied, which usually means the same platform that produced the first signature.

Encrypted documents cannot be opened for editing, so a password-protected PDF has to be opened first — see removing a known PDF password, and note that doing so produces an image-based copy with no form fields left to fill. In practice, if you need to fill a protected form, ask the sender for an unprotected copy rather than unlocking one yourself.

Fields the document's author marked read-only are skipped rather than overwritten, which is deliberate. Field appearances are regenerated in a standard font, so a filled form may not match the typography the author intended. Everything runs in your browser: the document and your signature are not transmitted anywhere, and the saved file is checked to be a structurally valid PDF before you are given it.