A Drawn Signature Is a Real Signature — Just Not a Qualified One

Three tiers, and what separates them

The first tier is defined by exclusion: any data in electronic form attached to or logically associated with other data, used by the signatory to sign. A drawn squiggle, a typed name, a scanned image of your signature, a tick in a box — all of these qualify. The bar is deliberately low, because the point of the definition is to stop courts from throwing out electronic evidence on a technicality rather than to certify anything.

The second tier adds four requirements at once: the signature is uniquely linked to the signatory, capable of identifying them, created using data the signatory can use under their sole control, and linked to the signed data in a way that makes any subsequent change detectable. That last requirement is the one a drawn image can never meet, because an image sitting on a page has no relationship to the bytes of the page. The third tier takes an advanced signature and adds two institutional facts: it was created on a qualified device, and it rests on a certificate from a provider a member state has vetted and listed. Those two facts are what earn the equivalence to a handwritten signature, and they are exactly the parts that cannot be produced by software running on your own machine.

What a drawn signature actually is, in the file

It is a picture placed on a page. The tool embeds an image and draws it at coordinates you chose, exactly as it would place a logo or a stamp. Nothing about that operation binds the image to the content underneath it: the page can be edited afterwards, the figures changed, a clause replaced, and the signature will sit there looking identical because it never knew what it was sitting on. That is not a flaw in the implementation. It is what drawing means.

A cryptographic signature works the other way round. It computes a digest over the exact bytes of the document and encrypts that digest with a private key, so the resulting value is a statement about this file and no other. Change one character and the digest no longer matches, and every reader is told the signature is invalid. That is the property people imagine they are getting when they draw their name, and it is worth knowing they are not.

When a simple signature is the right answer

Most of the time. The volume of documents that need signing vastly exceeds the volume that will ever be disputed, and treating every internal approval as if it were a mortgage deed is a way to make nothing get signed at all. A drawn signature is proportionate wherever the parties already know each other, the amounts are ordinary, and the surrounding evidence is strong — a signed quotation returned by email from a known address, sitting in a thread that establishes who agreed what and when, is in practice extremely hard to walk away from.

When it is not, and no amount of care will help

Two situations change the answer completely. The first is where national law prescribes a form: the transfer of real property, certain acts in employment and family law, sworn declarations, and anything a notary must receive. There the requirement is not that a signature exist but that it be of a specified kind, and a simple signature does not become sufficient by being carefully made. The second is where you expect a dispute — a termination, a contested settlement, a counterparty you do not trust — because there the absence of a presumption is exactly what will cost you. Both are cases for a qualified signature from a trust service provider, obtained through their process rather than produced on your own device.

The practical trap: signing then editing

Because a drawn signature is only an image, any later operation on the document leaves it in place and unremarkable — merging the file into a pack, compressing it, splitting it, adding a page. The signature travels along, still looking valid, still saying nothing about what it now sits beside. Sequence matters more than it appears to: assemble the document completely, then sign it last, and if anything changes afterwards, sign the new version rather than reusing the old signature.

Frequently asked questions

Is a signature I drew on a PDF legally binding?

In most cases yes, because most contracts require no particular form and can be concluded by any means that shows agreement. The signature is evidence of that agreement, and Article 25(1) prevents a court from disregarding it merely because it is electronic. The qualification is that it proves less than you may think: it does not by itself establish who made the mark, and it says nothing about whether the document changed afterwards. Where the law prescribes a form, or where the amount makes a dispute realistic, use a qualified signature.

Does printing, signing and scanning make it stronger?

Not meaningfully. A scan of a wet signature is still an image on a page, and it is arguably weaker than a drawn one because it is trivially copied from any other document you ever signed. It costs a printer, a scanner and ten minutes, and buys a feeling rather than a legal property. If the document matters enough to justify the ritual, it matters enough to justify a qualified signature, which costs less time than the round trip through paper.

How do I get a qualified signature?

From a qualified trust service provider, and only from one. Each member state publishes a trusted list of the providers it supervises, and the European Commission aggregates those lists; a provider not on one is not qualified whatever its marketing says. The process involves identity verification, which is the part that takes time and is also the part doing the work. Expect to prove who you are once, then sign as often as you like. No tool you run yourself, including this one, can substitute for that step.

Does this apply outside the European Union?

The three-tier structure is European. Other systems reach a similar practical place by different routes: several common-law jurisdictions treat almost any electronic mark as sufficient for most contracts, with the burden falling on whoever asserts a signature to prove it. The stable lesson across systems is the one this article is built on — a mark on a page is evidence, not proof, and where the stakes justify it, use a method that binds the signature to the document's contents.

Someone sent me a PDF with a signature image. Should I accept it?

Usually yes, and keep the context that came with it. The email it arrived in, the address it came from, the thread that preceded it and the date are the evidence that turns an image into a signature you can rely on — none of which is in the PDF. Store the message alongside the document rather than filing the attachment alone. If the value at stake would make a dispute worth someone's while, ask for a qualified signature instead, and expect that request to be understood rather than resented.