How to sign a PDF on any device, without Adobe
What a drawn signature on a PDF actually is, when it is legally enough, when it is not, and how to place one without installing anything or uploading the contract.
Someone has sent you a document to sign and return. The path most people take is: print it, sign it, photograph it, and send back a crooked, shadowed picture of a page. It works, and it produces the worst-looking document either of you will handle this month.
There is a better route, and understanding what it actually does — and does not do — matters, because “signature” means two completely different things in this context and people routinely believe they have the second when they have the first.
Two things called a signature
A drawn or typed signature is an image placed on a page. It is a picture of your name in your handwriting, positioned where the signature line is. That is all it is. It carries no proof of who drew it, no timestamp and no protection against being moved to a different document.
A digital signature is cryptography. A certificate issued to you, tied to an identity someone verified, used to produce a hash of the document’s bytes that anyone can check. It proves both who signed and that not one byte has changed since. Alter a comma and it breaks — which is the point.
These are not two grades of the same thing. They are different mechanisms answering different questions. This article is about the first one, because it is what almost every request actually needs.
When the first kind is enough
More often than people assume. In many jurisdictions an electronic signature — including a drawn one — is legally valid for ordinary agreements, on the basis that what matters is demonstrable intent to be bound rather than the technology used to express it.
Where the law does draw lines, it tends to draw them by how much assurance a signature carries rather than by whether it counts at all: an ordinary electronic signature at one end, and at the other a certificate-based one issued against a verified identity, which some legal systems treat as equivalent to a signature in ink. A drawn image flattened into a page sits firmly at the first end. That is a description of the shape of the rules, not advice about your situation — the specifics differ by country and by document type, and this article is not the place to settle either. If something turns on the answer, ask someone qualified in the jurisdiction that governs the document.
In practice, a drawn signature is fine for a delivery note, an internal form, a school permission slip, a rental application, a supplier form, an NDA between parties who trust each other, and most of the paperwork that arrives by email asking for “a signed copy”.
It is not enough where the law or the counterparty demands more: property transfers, wills, certain notarised instruments, and anything a regulator will audit. It is also not enough when the other side has told you they need a certificate-based signature — in which case they will usually send you into a platform that handles it, and you should let them.
The honest rule: if the question “could this ever be disputed, by someone with a lawyer” has a plausible yes, you want the second kind and probably a service built for it.
What “flattening” means, and why it matters
Placing a signature can be done two ways, and only one of them is safe to send.
An annotation floats above the page as a separate object. It can be selected, dragged, edited or deleted by anyone who opens the file in the right software, and some viewers do not render annotations at all — so the recipient may receive what looks like an unsigned document.
Flattening draws the signature into the page content itself. It becomes part of the page, the same as any other graphic. It cannot be dragged off, and every viewer renders it because there is nothing special left to render.
Always send the flattened version. If a tool offers you the choice and you are not sure which you have, open the result and try to click on your own signature. If it selects, it is an annotation.
Making a signature that does not look like a screenshot
Two details separate a signature that looks placed from one that looks pasted.
Transparency. If your signature image has a white background, it will sit on the page as a white rectangle covering whatever was underneath — a signature line, a box border, the sentence above it. The image needs a real alpha channel and needs to be cropped tight to the ink. Photographing a signature on paper gives you neither, which is why photographed signatures look wrong.
Resolution. A signature drawn at screen size and scaled up on the page will be visibly soft against crisp vector text. Export at several times the size you will place it, so the pixels are smaller than the letterforms around them.
Drawing directly with a finger, a stylus or a trackpad avoids both problems, because there is no photograph involved — the signature pad exports transparent, cropped, and at whatever scale you ask for, in PNG for placing and SVG for anything that will be printed. Draw it slowly. Signature pads sample your pointer, and a fast scribble produces fewer samples and a polygonal, obviously-digital line.
Placing it on the document
The mechanics, in order:
- Open the PDF in the signing tool. The file is read in the tab.
- Draw, type or upload the signature. Typed is fine for a name field; drawn for anything that says “signature”.
- Drag it onto the signature line and size it to match the surrounding text. Too large is the common mistake and it reads as unserious.
- Add the date and any initials the document asks for, as separate placements.
- Flatten and download.
- Open the result and check every page. A signature placed on the wrong page, or half over the edge, is invisible to you and obvious to them.
If the document arrived as photographs rather than a PDF, convert it first and sign the assembled document, rather than signing one image and hoping the ordering survives.
Multi-page documents, initials and dates
Most real agreements want more than one mark, and the requests are usually implicit.
Initials on every page is a common instruction, and its purpose is to make page substitution detectable: if page four can be swapped without anyone noticing, the initials are what makes the swap visible. Place them in the same position on each page — the footer corner is conventional — rather than wherever the pointer happened to be.
The date field is usually typed rather than drawn, and it is worth writing it unambiguously.
03/04/2026 is two different dates depending on which side of the Atlantic the reader is on. Write
the month as a word.
Print name fields are asking for legibility, not for a second signature. Type them.
Blank fields you were not asked to fill should stay blank. Filling in a field the other party intended to complete themselves creates a document neither of you agreed to.
Before sending, go through the whole file once at full size and check three things: every mark you placed is where you think it is, nothing overlaps text underneath it, and no page is missing a mark the document asks for. A contract returned with page six unsigned comes straight back to you, and the second round of email costs more time than the check.
Why this should not involve an upload
Consider the category. The documents people sign are contracts, leases, employment agreements, financial forms and settlements — the small set of documents that combine a full legal name, an address, financial terms and, once you have signed, a specimen of your signature.
That last one is worth sitting with. A signature image is a reusable credential. It can be lifted out of a returned PDF and placed on a different document by anyone who has the file, which is precisely why “flatten and check what you send” is advice worth taking seriously — and precisely why uploading the signed contract and the signature image to a free web service is a poor trade for saving two minutes.
None of the work requires a server. Reading the page, drawing an image into its content stream and writing the file back out is arithmetic a browser does instantly. The tools here do it in the tab: nothing is transmitted, and any signature you save for reuse is stored by your own browser on your own device, where you can clear it like any other site data.
The five-minute version
Draw the signature once, transparent and cropped. Place it, flatten it, check every page, send it. Keep the unsigned original.
And if the document is one where a dispute is genuinely conceivable, stop and use a certificate-based service instead. Knowing which of the two you need is most of the skill here.