Add watermark to PDF
Stamp text or an image across every page. Configure position, opacity, rotation, and tiling — see every change rendered live before you apply.
Add watermark to PDF stamps text (CONFIDENTIAL, DRAFT, URGENT, your custom string up to 100 characters) or an image (PNG, JPG, WebP, SVG up to 5 MB) across every page using a real PDF overlay — no rasterisation, no upload, no daily limit. Pick from nine positions or tile across the entire page; configure opacity, rotation, size, and skip-first-N. Every change re-renders live on every page before you apply. The watermark image (often a company logo) and the source PDF both stay in your browser; nothing is uploaded.
What is Add Watermark to PDF?
Add watermark to PDF stamps a text label or an image (typically a company logo) across every page of your PDF, then hands you back a single PDF with the marks burned into the page bytes. The watermark is selectable text or embedded image — not a separate metadata layer — so it travels with the file and shows up identically in every PDF reader on every device.
Pick text or image, drop in your content, choose one of the nine standard positions (or tile across the entire page), set opacity / rotation / size — and watch every page update live. The default of 25% opacity at −45° in middle-center matches the canonical CONFIDENTIAL stamp and is the right starting point for most documents; the modifiers are there for the cases where it isn't.
How does it work?
The watermark is drawn as a real text or image overlay on each page using pdf-lib. Text watermarks use one of three standard PDF fonts (Helvetica for Sans, Times Roman for Serif, Courier for Mono) — no extra fonts are embedded in the file. Image watermarks are embedded directly as PNG or JPG; SVG is rasterised to PNG at 2× resolution first, since the PDF format doesn't support SVG natively. The original page content is preserved underneath the watermark — the marks live on top of the existing page bytes, not as replaceable metadata.
Text vs image watermark — when to use each
Text watermarks are for status stamps — CONFIDENTIAL, DRAFT, REJECTED, URGENT, VOID, NOT FOR DISTRIBUTION. They communicate one piece of information loudly and don't depend on any visual asset. Image watermarks are for branding — company logos on client deliverables, certification badges on credentials, charity / NGO marks on advocacy documents, photographer / artist marks on portfolios. The opacity guideline is different for each: text watermarks work best at 25–50% (visible but readable through), image watermarks at 15–30% (logos sit behind text content rather than competing with it). Both can tile across the page when you want maximum visual presence, like a high-security document or screenshot-prevention overlay.
When should I use tile mode?
Tile mode repeats the watermark in a 3 × 4 grid across each page, replacing the single-position layout. Use it for high-security documents where you want every region of every page marked (so screenshots can't easily exclude the watermark), for brand reinforcement on a leave-behind document, or for screenshot-prevention overlays where the watermark intentionally interferes with content extraction. Single-position is right for everyday use — status stamps, footer notices, page-corner branding — and the default 25% opacity at middle-center −45° is the canonical CONFIDENTIAL stamp shape that's been in use for over a century.
Can recipients remove the watermark?
Watermarks added by this tool — and by every non-cryptographic watermarking tool, Smallpdf and iLovePDF included — can be removed by anyone with PDF editing software. They sit as a normal page overlay on top of the original page content; Acrobat Pro, PDF24, and a handful of free tools can select and delete them in seconds. This isn't a flaw — it's how visible watermarking works. Watermarks are designed for visible deterrence and clear status marking (CONFIDENTIAL, DRAFT, NOT FOR DISTRIBUTION) — they communicate intent loudly, not as a security control. For tamper-resistant marking — where the recipient cannot easily strip the mark — you need a digital signature, which is cryptographic. A Sign PDF tool is on our roadmap for that use case.
Why does the watermark image privacy matter?
Image watermarks deserve a special note. Company logos, brand badges, and signature images are sensitive — they identify your organisation and, in the case of signatures, you personally. Many watermarking tools upload the watermark image to their server alongside the source PDF; some retain the image for caching or branding-detection purposes; some explicitly state in their terms that uploaded images may be used to improve their service. Pikowl's watermark image is decoded entirely in your browser via the canvas API, embedded by pdf-lib in your browser, and never sent anywhere. The same is true of the source PDF. Open your browser's Network tab while applying a watermark and you'll see thumbnail and worker chunks load — but no upload of either your PDF or your image.
When would I use this?
The most common cases are status stamps on draft documents (CONFIDENTIAL, DRAFT, NOT FINAL) for internal review, branding stamps on client-facing deliverables (PNG company logos placed unobtrusively in a corner), VOID / REJECTED stamps on documents that have been superseded but need to remain on file, embargoed-press marks on pre-publication content, and tiled-watermark protection for high-security documents where screenshots and partial prints both need to carry the mark.
What are the limits?
Per-file limit is 100 MB and 500 pages for the source PDF. Watermark images are capped at 5 MB and must be PNG, JPG, WebP, or SVG. Watermark text is capped at 100 characters — long enough for any real status stamp (CONFIDENTIAL FOR INTERNAL USE ONLY — DO NOT DISTRIBUTE fits comfortably) but short enough that tile mode renders cleanly. Above 30 pages, the live preview shows the first 6 and last 6 pages plus a "+ N more pages" indicator; the actual watermark still applies to every non-skipped page.
How does Pikowl compare to Smallpdf and iLovePDF?
How Pikowl's add-watermark compares to two of the most common alternatives. The headline differences are the live preview, the tile mode controls, and that the watermark image (often a company logo) never leaves your device.
| Feature | Pikowl | Smallpdf | iLovePDF |
|---|---|---|---|
| Processing | 100% local — both source PDF and watermark image stay in your browser | Both source PDF and watermark image uploaded to a server, then deleted afterwards | Both source PDF and watermark image uploaded to a server, then deleted afterwards |
| Per-file limit | Up to 100 MB / file, 500 pages, 5 MB watermark image | Up to ~5 MB free; larger files require Pro | Up to ~10 MB free; larger files require Premium |
| Live preview | Live preview re-renders every page on every settings change | No live preview — settings apply only on download | Static preview only — no live re-render on settings change |
| Tile across page | Yes — single toggle, automatic 3×4 grid spacing | Not available | Available, but settings less granular |
| Pricing | Free, no daily cap | Free with 2 tasks/day cap; Pro $12/mo | Free with daily cap; Premium $7/mo |
Last updated April 2026. Smallpdf and iLovePDF figures from their respective free-tier specs at time of writing.
Common Uses
- Marking documents CONFIDENTIAL or DRAFT before sharing: The default tool — text watermark, middle-center, 25% opacity, −45°. Lets recipients see the document is sensitive without obstructing the content.
- Adding a company logo to client deliverables: Drop your PNG logo, place it bottom-right or middle-center at low opacity (15–25%), so the document is clearly branded without competing with the content.
- Stamping VOID or REJECTED on a contract that's been pulled: Red text watermark, large size, −45° diagonal across center — communicates the status at a glance without rewriting the document.
- Watermarking photo prints or design exports inside a PDF: Tile a small logo or text across the entire page so any screenshot or partial print still carries the mark.
- Adding a charity / NGO mark to an advocacy report: Image watermark at the top of every page, low opacity, marks ownership without distracting from the content.
- Pre-publication manuscript stamps: EMBARGOED UNTIL [DATE] in red text, tiled across every page — clear, hard to miss, deters early sharing.
Frequently asked questions
Can I rotate the watermark to any angle?
Three preset angles — −45° (the canonical diagonal stamp), 0° (horizontal), and +45° (diagonal in the other direction). Arbitrary rotation in single-degree increments isn't in v1; we've found the three presets cover ~95% of real watermark needs (CONFIDENTIAL diagonal, header / footer text, and the occasional reverse-diagonal). Free-input angles are on the roadmap if usage shows real demand.
Will the watermark cover up my text?
It will if you crank opacity to 100%. The default of 25% leaves the underlying text fully readable while still making the watermark obvious — that's the sweet spot for most documents. For very dense pages (small font, lots of figures), drop to 10%; for cover pages where the watermark is the focus, raise to 50%. The live preview shows you exactly what overlap will happen on every page before you apply.
What image formats can I use as a watermark?
PNG, JPG, WebP, and SVG up to 5 MB. PNG is the best choice for logos with transparency. JPG works for photographic watermarks (badges with backgrounds, stamps with shading). WebP is supported and rasterised to PNG at upload time. SVG is rasterised to PNG at 2× the intrinsic resolution before embedding, so it stays crisp at typical watermark sizes — if your SVG looks soft in the output, increase its intrinsic dimensions in the source file rather than relying on browser scaling.
Can the watermark be removed by recipients?
Yes — and it's important to be honest about this. Watermarks added by this tool (or any non-encrypted PDF watermarking tool, including Smallpdf and iLovePDF) sit as a normal page overlay on top of the original page content. Anyone with PDF editing software (Acrobat Pro, PDF24, etc.) can select and delete them. For tamper-resistant marking — where the recipient can't easily strip the mark — you need a digital signature, which is cryptographic rather than visual. We're planning a Sign PDF tool for that use case. Watermarks are useful for visible deterrence and as a clear status marker (CONFIDENTIAL, DRAFT, NOT FOR DISTRIBUTION) — that's what they were designed for, not for tamper-resistance.
Why does my watermark image look fuzzy?
Most likely your source image is low-resolution. The watermark in the output PDF is rendered at the size you picked (15 / 30 / 50 / 75% of page width); if your source PNG / JPG is only 200 pixels wide and you're rendering it at 75% of an A4 page, the result will look pixellated. Use a source image at least 1500 pixels wide for crisp output at any size. SVG sidesteps this — it always rasterises at 2× the intrinsic dimensions, so you can scale freely.
Are my files ever sent to a server?
Yes. The whole tool runs in your browser using pdf-lib for the PDF rendering and pdf.js for the preview thumbnails. Neither the source PDF nor the watermark image (often a company logo with branding you may not want on third-party servers) is ever uploaded. Verify by opening the browser's Network tab while you process a PDF — you'll see thumbnail and worker chunks load, but no upload of your file content.
What are the file-size limits?
Up to 100 MB per source PDF, 500 pages per document, 5 MB per watermark image, 100 characters per watermark text. The 5 MB watermark-image cap keeps preview rendering fast and prevents giant images from bloating the output PDF; in practice, a well-prepared logo PNG is typically 50–500 KB and an SVG can be a few KB. Above 30 pages, the live preview shows only the first 6 and last 6 thumbnails to keep the panel responsive — the actual watermark application still runs on every page.