Add watermark to PDF for free — text or image

What a watermark actually does

A watermark is a low-tech, high-effect signal layered on top of page content. Three things it is useful for and one thing it is not.

• State labelling. The DRAFT / CONFIDENTIAL / REVIEW COPY tag is the most common case. It tells anyone looking at the document — including the person who copied it onto a tablet, printed it, forwarded it twice, and finally noticed the corner stamp — that this version is not the canonical one. The stale-draft-treated-as-final mistake is one of the most common accidental disclosure modes in business; a watermark makes it hard to make.
• Provenance / attribution. A small footer watermark with "© 2026 Company Name" or a logo on every page identifies the document source so a snippet shared out of context can still be traced back to you. Photographers and illustrators have done this on images for centuries; the same principle applies to PDFs.
• Leak attribution (forensic watermarks). A per-recipient unique watermark (different ID for each copy distributed) lets you identify which recipient's copy ended up somewhere it should not have. Not currently a one-click feature in the ScoutMyTool tool — but trivially scriptable by running the text watermark in a loop with a different text value per recipient.
• NOT useful for: preventing recipients from removing the content. A watermark is part of the rendered page, not a permission flag. A sufficiently determined recipient can rasterise the page, in-paint over the watermark, and re-OCR the text. For "the recipient genuinely cannot extract this content", you need DRM (digital rights management) — a different category of tool, not free, and rarely what the average user actually needs. Watermarking is a signal, not a lock.

How to add a text watermark

The ScoutMyTool text watermark tool lives at scoutmytool.com/pdf/add-watermark (a common search query is the short form scoutmytool.com/pdf/watermark-pdf — the canonical URL is the "add-watermark" one). Runs entirely client-side via pdf-lib.

1. Open the tool and drop your PDF. One file at a time. Loaded into a sandboxed memory buffer; nothing is uploaded.
2. Type the watermark text. Default is "CONFIDENTIAL". Other common labels: DRAFT, REVIEW COPY, DO NOT COPY, INTERNAL USE ONLY, FOR DISCUSSION, NOT FOR DISTRIBUTION. Maximum 80 characters. Latin / numeric characters only in the current version — for non-Latin scripts, use the Image Watermark tool with a pre-rendered PNG.
3. Pick the position. Three options:
   • Diagonal across the page (most common) — 45° rotation, centred. The classic DRAFT-stamp look.
   • Horizontal centre — same centred position but un-rotated. Useful when you want a less-aggressive stamp.
   • Small footer right — bottom-right corner, smaller font, useful for copyright / attribution lines that should not obscure body text.
4. Set the font size. Default 60pt. Guidance:
   • 40–80pt — typical for diagonal / horizontal centre on a US-letter page. Large enough to be unmissable, small enough not to bleed off the page.
   • 8–14pt — typical for footer attributions. Visible but not dominating.
   • 100–200pt — for posters, A0+ documents, or when you really want the message to be the first thing anyone sees.
   The tool clamps to 6–200pt; values outside that range are treated as the nearest valid value.
5. Set the opacity. Default 25%. Lower = more subtle. 20–30% is the "background watermark" sweet spot — clearly visible but does not obscure the body text underneath. Bump to 50% if you want it more prominent; drop to 10% if you want it barely there. The tool clamps to 5–100%.
6. Click "Add Watermark". The tool walks every page, draws your text at the chosen position, opacity, and size, and saves the result as <your-pdf-name>-watermarked.pdf. On a typical 30-page document the entire operation finishes in well under a second.
7. Verify before sending. Open the output, scroll through. Check that the watermark appears on every page, in the right position, at the intended opacity. If something is off, the source PDF is untouched — re-run with adjusted parameters.
8. If your PDF is password-protected. The tool refuses encrypted sources with a clear error message. Unlock first via Unlock PDF, watermark the unlocked copy, then re-protect via Protect PDF if needed.

How to add an image watermark (a logo)

For logos, scanned signatures, or any image-based watermark, use the separate Image Watermark tool. Drop one PDF and one image, choose options, click. Steps:

1. Prepare the watermark image. PNG with a transparent background is strongly preferred — the underlying page content shows through where the logo is transparent, giving the classic "logo overlay" look. JPG works but has no transparency, so the logo will appear inside an opaque rectangle. Pre-scale to no more than ~1200 pixels wide; high-resolution logos embedded into every page can balloon the output file size.
2. Drop both files together. One PDF and one image, in either order. The tool detects which is which automatically from file extension and MIME type.
3. Pick the position. Five options: Center, Top Right, Top Left, Bottom Right, Bottom Left. "Center" is the full-page diagonal-style watermark; the four corner positions are for small attribution marks.
4. Pick the scale. 15% / 30% (recommended) / 50% / 80% of page width. 30% is the right starting point for most logo overlays; bump to 50% for a more prominent stamp.
5. Set the opacity. 5–90%. Default 25%. Same guidance as for text watermarks.
6. Click "Image Watermark". The image is embedded into the PDF object structure and stamped on every page at the chosen position. PNG transparency is preserved; the underlying page content shows through transparent regions.

How ScoutMyTool compares to Smallpdf, iLovePDF and PDF2Go

All four offer watermarking on the free tier — but with different free-tier caps, different default-paywall behaviour, and different answers to "is my file uploaded".

Third-party gating, free-tier caps, and feature availability taken from each vendor's public product pages as of May 2026 and may change.

A note on document-class standards and watermarks

PDF/A is the ISO-standardised archival variant of PDF (ISO 19005, most recent revision ISO 19005-3:2012)¹ designed for long-term preservation of electronic documents. Watermarks added to a PDF/A file should themselves use only embedded fonts and remain valid against the conformance level (PDF/A-1, PDF/A-2, PDF/A-3) of the source — the ScoutMyTool tool embeds the standard Helvetica font subset, which is conformant for PDF/A-2u and PDF/A-3u (the most common variants used today for legal and government archival).

For legally-binding documents intended for court submission, the E-Discovery industry recommendation is that watermarks be visible, unambiguous, and embedded into the page content stream (not as removable annotations) — the U.S. Sedona Conference Cooperation Proclamation guidance on production formats explicitly notes that annotation-only watermarks are insufficient for production copies². The ScoutMyTool tool draws watermarks into the content stream, which satisfies this requirement.

Related PDF tools on ScoutMyTool

• Add Watermark to PDF — text watermark, 3 positions, customisable size + opacity.
• Image Watermark — logo / image stamp, 5 positions, scale + opacity.
• Sign PDF — drop a signature image (a specialised form of image watermark).
• Protect PDF — pair watermarking with encryption for "draft + locked".
• Permanent Redaction — for content that genuinely needs to be unrecoverable, not just watermarked.
• Merge PDF — combine watermarked subsets with un-watermarked content.
• Extract Pages — isolate pages before watermarking specific ones only.

Frequently asked questions

What watermark text and position should I pick?

For "this is a draft" or "do not distribute" labelling, the standard is a diagonal 45-degree watermark across the page centre, font size 60–80pt, opacity 20–30%, all-caps grey text — DRAFT, CONFIDENTIAL, DO NOT COPY, REVIEW COPY are the most common labels. For a more subtle copyright / attribution mark, use the small footer-right position at 8–10pt with full opacity. For a centred "header"-style label (less common in 2026), pick horizontal centre at 40–60pt and 30–50% opacity. The diagonal default is what every "draft watermark" example you have ever seen looks like, and it is the right starting point for 80% of use cases.

Can I add an image watermark (a logo) instead of text?

Yes — use the separate Image Watermark tool. Drop one PDF and one image (PNG with transparency works best, but JPG also fine), pick a position (center / topright / topleft / bottomright / bottomleft), pick a scale (15% / 30% / 50% / 80% of page width), set opacity, and click. The image is stamped on every page at the chosen position. PNG with transparency is recommended because the underlying page content shows through; JPG works but adds a visible rectangle of white around the logo where the JPG has no alpha channel.

Is the watermark permanent, or can recipients remove it?

The watermark is drawn into the page content stream as a normal text or image object — not as a removable annotation. It survives printing, re-saving in another reader, copying-and-pasting into a new file, and screenshotting. A sufficiently determined attacker can still rasterise the page, in-paint the watermark, and re-OCR — for genuinely irreversible protection of sensitive content, use Permanent Redaction (which rasterises the redacted regions) rather than a watermark. For the everyday "DRAFT — do not circulate" use case, the standard watermark is the right answer; for "I genuinely need the recipient unable to remove this", you are in the wrong tool category and should look at DRM.

Will the watermark appear on every page, or can I target specific pages?

The text watermark applies to every page in one click — no per-page selection. If you need to watermark only specific pages (e.g. only the cover page, or only the appendix), use a two-pass workflow: run Split PDF or Extract Pages to isolate the pages you want to watermark, run them through Add Watermark, then merge the watermarked subset back with the un-watermarked rest via Merge PDF. The image watermark behaves the same way — every page or none.

Is my PDF uploaded to your server?

No. The text watermark tool runs entirely in your browser using pdf-lib. Your PDF is loaded into a sandboxed memory buffer, the text is drawn into each page's content stream, and the watermarked output is delivered as a download. Same for the image watermark tool — both the PDF and the image stay on your device. Verify in the browser network tab.

Can I use non-Latin characters in the watermark text?

Not in the current version. The text-watermark tool uses the built-in Helvetica font which only encodes WinAnsi (Latin-1) — Arabic, Hebrew, Greek, Cyrillic, and CJK characters fail with a clear pre-flight error rather than rendering as garbage. TrueType embedding for non-Latin scripts is on the roadmap. Workaround for now: use the Image Watermark tool — render your non-Latin text as a PNG in any drawing app (with transparency), then drop the PNG as a watermark image.

How big can the PDF be?

No hard cap — the text watermark runs client-side, so the only limit is your device's RAM. Practical limit is around 500 pages or 100 MB on a typical laptop. The image watermark also runs client-side and has the same effective limit, plus an additional consideration: a high-resolution logo embedded into every page can grow the output file noticeably. Pre-scale the logo to a reasonable resolution (a 1200-pixel-wide PNG is plenty for any "30% page width" stamp) before uploading.