Two quite different things share the word signature, and the gap between them occasionally matters a great deal.
One is an image of your handwriting placed on a page. The other is a cryptographic operation that ties a specific document to a certificate held by a specific person, and which fails visibly if a single byte of that document changes afterwards.
What each one actually proves
| Image of a signature | Cryptographic signature | |
|---|---|---|
| Proves who signed | weakly | strongly, via a certificate |
| Detects later edits | no | yes, immediately |
| Reusable by anyone holding it | yes | no |
| Accepted for most forms | yes | yes |
| Accepted where law requires it | often not | yes |
The reusability line is the one worth pausing on. An image of your signature is a file. Anyone who receives a document containing it can extract it and place it on something else, and nothing about the result would look wrong.
That is not a reason never to use one — it is a reason to treat the file the way you would treat a physical rubber stamp, rather than as something to email around.
When is an image enough?
Most of the time, in practice. Internal approvals, delivery notes, school forms, letting agreements, the endless PDFs that ask for a signature because the paper version had a line for one.
The legal weight is roughly that of a signature on a fax: it demonstrates intent, and in most common-law jurisdictions intent is what a signature is for. Disputes about them turn on whether the person meant to sign, not on the pixels.
When is it not?
Where a statute names a standard. The EU eIDAS regulation defines three tiers — simple, advanced and qualified — and certain transactions require a qualified electronic signature, which needs a certificate issued by a trust service provider after identity verification. An image satisfies none of that.
Property transfers, wills, and some financial instruments carry their own requirements which vary by jurisdiction and are worth checking rather than assuming.
Why transparency matters for the image
A signature saved on a white background is a white rectangle. Place it on a form that is not pure white — a scanned document, a coloured letterhead, a PDF with a tint — and the rectangle is visible. It looks pasted on, because it is.
A PNG with a transparent background has no background at all, so only the ink lands on the page. It is the single difference between a signature that looks placed and one that looks stuck on.
Why draw rather than type?
Both work, and they suit different situations.
A drawn signature looks like yours, which matters when it sits beside a printed name on a document someone will glance at. Drawing with a trackpad produces something wobblier than your pen version, and a finger on a phone screen is usually closer.
A typed signature in a script font is legible and consistent and looks like what it is. For internal forms nobody minds; for anything a client sees, the drawn version reads better.
Questions people ask
Can I just paste a photo of my signature on paper? You can, and it will carry a grey or white background and the texture of the paper. Drawing directly gives a clean transparent result.
Does the signature need to match my passport? For an image, no — there is no comparison being performed. That is rather the point of the distinction.
Is a typed name at the bottom of an email a signature? Courts have found that it can be, where intent is clear. That is a question about intent, not about formatting.
How do I stop mine being reused? Send the flattened document rather than the signature file, and do not keep it somewhere shared.
The signature generator draws one and saves it as a transparent PNG without uploading anything, and the certificate maker and business card maker produce the documents it tends to end up on.