What a PDF watermark is actually for
A watermark on a PDF is a piece of text or imagery stamped across every page that says something about the state of the document. The four words that show up on 90% of watermarks are DRAFT, CONFIDENTIAL, SAMPLE, and SPECIMEN. The next most common are a person's name, a date, or a company name used as a soft "do not redistribute" marker.
The point of a watermark is visual context, not security. When someone opens a contract and sees "DRAFT" diagonally across every page, they know not to sign it or quote from it. When a financial report has "CONFIDENTIAL" stamped across every page, anyone who screenshots it leaves the marker in the image. The watermark doesn't stop the screenshot — it makes the screenshot self-incriminating.
The Microapp PDF Watermark tool stamps custom text across every page of a PDF, with controls for color, opacity, rotation, and font size. The whole operation runs in your browser via pdf-lib. The file doesn't upload. Your draft contract doesn't sit on a server somewhere while you make coffee.
How to use the PDF Watermark tool
One screen, five inputs, one button. Drop a PDF, type the text, tune four sliders, click.
- Drag a PDF onto the drop zone, or click to pick one. Up to 100 MB.
- Type the watermark text. Up to 60 characters.
DRAFT,CONFIDENTIAL,SAMPLE, your full name, today's date — anything that reads at a glance. - Pick a color. Grey is the default and looks the most like a real watermark. Red is the loudest. Blue and black are alternatives.
- Set font size (default 72pt), opacity (10-100%, default 30%), and rotation (0° horizontal, 45° diagonal, 90° vertical). The defaults match the classic diagonal-DRAFT stamp.
- Click Add watermark. Output is named after the file plus the text —
contract.pdfwith textDRAFTbecomescontract-draft.pdf.
The watermark is centered on every page. There's no per-page positioning in v1 — if you need that level of control, you're building a layout, not stamping a marker.
Settings that produce specific looks
The defaults — grey, 72pt, 30% opacity, 45° rotation — produce the diagonal watermark you've seen on a thousand printed drafts. Different combinations produce different looks; here's the cheat sheet.
| Effect | Color | Size | Opacity | Rotation |
|---|---|---|---|---|
| Classic diagonal DRAFT | Grey | 72pt | 30% | 45° |
| Loud, hard-to-miss | Red | 96pt | 50% | 45° |
| Subtle background mark | Grey | 72pt | 15% | 45° |
| Horizontal banner across the top | Grey or red | 24pt | 40% | 0° |
| Vertical spine marker | Grey | 48pt | 30% | 90° |
| Confidentiality stamp | Red | 60pt | 40% | 45° |
If you can't decide: start with the defaults. They look right on a letter-size or A4 document about 95% of the time. Only tune up to a louder setting if you've previewed the output and it doesn't read forcefully enough; only tune down if it's fighting the body text.
A worked example: stamping a contract draft
You've written a 14-page services agreement. The client wants to see it before legal review. You want to send it without any risk of it being mistaken for the signing version.
In the tool:
- Text: DRAFT - NOT FOR SIGNATURE
- Color: Red
- Font size: 60pt
- Opacity: 35%
- Rotation: 45°
Click Add watermark. Every page of the agreement now has a red diagonal "DRAFT - NOT FOR SIGNATURE" stamped across the middle. Translucent enough that the contract text is fully readable; loud enough that nobody is going to print the file and sign it by accident. The whole operation takes about ten seconds. You email services-agreement-draft.pdf to the client with confidence.
Pair this with Add Header & Footer for documents that need both a watermark and pagination — for example, a draft contract with "DRAFT" diagonally and "Page X of Y | Draft 2026-05" in the footer. Two passes, same browser, still no upload.
The honest section: watermarks are not security
This needs to be said clearly because the iLovePDF and SmallPDF marketing copy doesn't say it. A text watermark added to a PDF is a drawing on top of the page. It is not encryption, not access control, not DRM. Anyone with a copy of Adobe Acrobat can open your watermarked file, click the watermark text, and delete it. Anyone with the free pdf-lib library can write fifteen lines of JavaScript that does the same thing in bulk. Watermarks deter casual misuse and signal intent. They do not prevent a determined actor from removing the marker.
If your document actually needs to stay confidential, the right tools are PDF encryption (a password to open the file), permission restrictions (a password to edit or copy from it), and redaction of sensitive content before distribution. The PDF Redact tool covers the last one — it removes content from the underlying PDF rather than hiding it under a black bar. For real encryption, use a desktop tool like Acrobat or qpdf.
That said, watermarks are great for what they're actually for: communicating state. "This is a draft." "This is internal." "This is a sample, not a final deliverable." "This is mine, please don't redistribute." Those messages get across, and the marker survives screenshots, prints, and email forwarding because it's drawn into the page rather than attached as metadata.
What about image watermarks?
v1 ships text watermarks only. We thought about including PNG/JPG logo watermarks and decided to ship the simpler thing first. The reasoning: text covers 90%+ of the use cases people actually have for watermarking a PDF (DRAFT, CONFIDENTIAL, name, date). Image watermarks add three new decisions — positioning, sizing, dealing with PNG transparency on a colored page background — that need their own UI and their own preview.
If you genuinely need a logo watermark today, the workaround is to make a one-page PDF that's just your logo on a transparent background, then use a desktop tool that overlays PDFs (Acrobat, PDFsam) to stamp it across your document. We'll likely add native image watermarks to the tool when it makes sense, but text is the right place to start.
Compared to iLovePDF, SmallPDF, PDF24, and Sejda
These four are the dominant Google results for "PDF watermark." Each takes a different shape of the same model: upload your file, watermark on the server, download or pay.
- iLovePDF caps free use at three files per day per browser. Watermark options are decent — text or image, with rotation and opacity — but everything runs on their server. Your draft contract sits in their queue.
- SmallPDF caps free use at two files per day and shows ads. Same server upload model. Their watermark UI is slightly cleaner; the privacy story is identical.
- PDF24 has a desktop client that works locally, but the web tool uploads. Mixed model, free, ad-supported.
- Sejda is the most powerful watermark UI of the four — per-page positioning, multiple watermarks, image support. Free tier caps at three tasks per hour and 200 pages.
The Microapp version trades off: less customization than Sejda, no image watermark in v1, no per-page positioning. In exchange, the file never leaves your machine, there's no daily cap, no ads in the tool, no signup, no upload progress bar. Pick the right one for the job — if you need image watermarks with pixel-level positioning, Sejda is genuinely the better tool. If you need to stamp DRAFT across a contract without uploading it, this is it.
What's preserved in the output
pdf-lib adds a single text element to each page. It does not re-render or re-encode anything else. That means:
- Bookmarks and TOC links survive.
- Hyperlinks in the body still work.
- Form fields stay fillable.
- Digital signatures will be invalidated, because any modification breaks the signature chain. This is unavoidable for any tool that modifies a signed PDF.
- File size grows by a small amount — typically a few KB total for the watermark text and embedded font, no matter how many pages.
If you want to add a watermark plus page numbers, run this tool first, then Add Page Numbers on the output. Both run in your browser; both leave bookmarks and links untouched.
Frequently asked questions
Is my PDF really not uploaded?
Correct. The watermarking runs in your browser via the pdf-lib JavaScript library. Open your network tab during the stamp operation and you'll see zero outbound requests. Your file goes from your disk to browser memory to the downloaded output, with no server in the loop.
Can I use a logo image as the watermark?
Not in v1 — only text. Image watermarks are on the roadmap but the UX (sizing, positioning, dealing with PNG transparency on different backgrounds) is meaningfully more complex than text and we'd rather ship the simple thing well than the complex thing badly. For now, text covers most of what people actually need.
Can I watermark only some pages, not all?
Not in v1 — every page gets stamped. To watermark a subset (say, everything except the cover page), use Split PDF to extract the pages you want watermarked, run them through this tool, then merge the cover back in.
Is the watermark removable?
Yes, technically. The watermark is drawn as a text element on each page; anyone with Adobe Acrobat or another PDF editor can select and delete it. The watermark deters casual misuse and survives screenshots and prints — it does not prevent a determined actor from removing the marker. If you need real document security, use PDF encryption and access control.
Can I have multiple watermarks on the same page?
Not in a single pass — but you can run the tool more than once on its own output. Each pass adds another watermark independently. So you could stamp "DRAFT" diagonally in red, save the output, then stamp the date horizontally in grey at the top.
Will it work on password-protected PDFs?
No. pdf-lib refuses to open encrypted PDFs. Remove the password first in a desktop reader, then run the watermark.
What's the maximum file size?
100 MB. The operation is fast because pdf-lib doesn't re-render pages — it just adds a text element per page. A 200-page PDF watermarks in under a second.