Add page numbers to a PDF — custom position and format

Step-by-step: number a PDF the way the document actually needs

The ScoutMyTool tool lives at scoutmytool.com/pdf/add-page-numbers. Runs client-side via pdf-lib — no upload, no signup, no quota.

1. Drop your PDF. One file at a time. The file loads into a sandboxed memory buffer — confirm in DevTools Network if the document is sensitive.
2. Pick the position. Six options — top/bottom × left/center/right. The default is bottom-center, which is the most common convention for printed reports. Use bottom-right for documents that will be photocopied (the left edge is more likely to be under the stapler / hole-punch). Use top-right for legal-style memos where the signature block lives at bottom-center.
3. Pick the format. Four options:
   • Numeric (1, 2, 3, ...) — the default.
   • "Page X of N" — the right answer when the document is meant to be printed and the reader needs to know the total. N is computed from the numbered region, not the raw page count, so excluding the title page produces "Page 1 of 23" not "Page 1 of 24".
   • Roman (i, ii, iii / I, II, III) — front matter convention. Combine with a second pass for Arabic body numbers.
   • Letter (a, b, c) — useful for appendix-style enumeration when the body of the document uses Arabic and the appendix should be visibly different.
4. Set "Pages to number". Range syntax: 2- means "from page 2 to the end" (skip cover), 3-12, 14- means "pages 3 through 12 and 14 onwards" (skip pages 1, 2, 13). The page-range syntax is the same as in Extract Pages and works the same way.
5. Set "Start at". Default 1. Set higher if the PDF is a fragment of a larger document — e.g. chapter 4 of a book physically starts as page 1 of its own file but should read "87" onward. The starting number applies to the first numbered page after any skipped front matter.
6. Set font and size. Helvetica 10 pt is the default. Times 11 pt matches a Word default. Courier 10 pt works for code-style documents. The number is added as real text (selectable, searchable, scalable), not rasterised.
7. Click Add Numbers. Each chosen page gets a new text object at the configured position; the rest of the PDF is left untouched. Operation is bytecode-level — output preserves all original fonts, images, and form fields.
8. Two-pass workflow for mixed numbering. Add lower-Roman to the front matter, save, reload the output, then add Arabic to the body. Both passes run client-side.
9. Verify in your reader. Open the output in Adobe Reader or Preview, scroll through, and check that the skipped pages are unnumbered and the numbered range starts at the right value. Print Preview is the surest way to confirm margin clearance against your printer's unprintable area.

How ScoutMyTool compares to Smallpdf, iLovePDF and PDF2Go

All four offer page-number stamping. The meaningful differences are around control: how many positions, which formats, whether you can skip specific pages, and whether the file leaves your device.

Third-party gating and feature claims taken from each vendor's public pricing pages as of May 2026.

Why "Page X of N" requires more thought than it looks

The naive implementation of "Page X of N" sets N to the raw page count of the input file. That produces wrong results whenever the document has front matter that should not be numbered (covers, table of contents, copyright pages). The reader of a 24-page report with a one-page cover expects to see "Page 1 of 23" on the first body page, not "Page 2 of 24" — because to the reader, the numbered region IS the document.

ScoutMyTool computes N as the count of pages that will receive a number, derived from the "Pages to number" range. If you set range "2-" on a 24-page PDF, N becomes 23 (pages 2 through 24), and the first numbered page reads "Page 1 of 23". If you set range "3-, 1" (cover at the end, body from page 3), N becomes 22 + 1 = 23 and the first body page still reads "Page 1 of 23". The convention is encoded in the PDF specification by reference via the document outline / page label feature (ISO 32000-1 §12.4.2)¹, which standardises how different page-numbering schemes (Arabic, Roman, Letter) can be applied to disjoint ranges of a single document.

Related PDF tools on ScoutMyTool

• Add Page Numbers — the tool this guide is about.
• Extract Pages — pre-filter to a subset before numbering.
• Merge PDF — combine multiple PDFs before numbering the combined output.
• Bates Numbering — the legal-discovery variant of page numbering.
• Watermark PDF — same general "stamp text onto every page" mechanism, applied to confidentiality marks rather than page numbers.
• PDF Editor — for manual placement of numbers in unusual positions the automated tool does not support.
• Unlock PDF — required first if your source is password-protected.

Frequently asked questions

Can I start numbering from a number other than 1?

Yes. The "Start at" field accepts any positive integer. The most common reason is when the PDF is part of a larger compiled document — for example chapter 4 of a book PDF that physically starts at page 1 of its own file but should print page numbers 87 onward. Set Start at = 87 and the first page of the PDF carries "87", the second "88", and so on. You can combine this with "Pages to number" to skip a title page or table of contents and still have the first numbered page show 1 (or any other value).

How do I get "Page 1 of 24" instead of just "1"?

Select the "Page X of N" format from the format dropdown. The tool computes N as the count of pages that will receive a number (so if you exclude the title page via "Pages to number = 2-", N becomes total pages minus 1, not the raw total). This is the right behaviour for documents like reports and theses where the reader expects "Page X of N" to refer to the numbered region, not the cover-inclusive total.

Where exactly does the number sit on the page?

Six position options: top-left, top-center, top-right, bottom-left, bottom-center, bottom-right. Margin is configurable in points (default 24 pt ≈ 8 mm), which keeps the number safely inside the visible area after typical print binding. The number is drawn with pdf-lib as a real text object — not rasterised onto the page — so it remains searchable, selectable, and scales perfectly when zoomed. You can pick the font (Helvetica, Times, Courier) and the font size (default 10 pt).

My PDF has a different page size on some pages — does the number stay in the corner?

Yes. The tool reads each page's actual mediabox dimensions and positions the number relative to that page's top/bottom and left/right edges. A mixed-size PDF (e.g. a Letter cover bound with A3 fold-out tables) ends up with correctly positioned numbers on every page, not "all positioned relative to the cover" as some lesser tools produce. You can verify by zooming into a fold-out page after numbering — the number sits the configured margin from that page's edge.

Will the page numbers replace existing page numbers in the document?

No. The tool draws on top of whatever is on the page — it does not parse or remove existing text. If your PDF already shows "Page 12" in the footer (drawn by Word or InDesign on export), and you add a number with this tool, you end up with TWO numbers in the footer. To clean up the original, either re-export from the source application without numbers, or use PDF Editor to redact the existing number area before numbering.

Is my PDF uploaded to your servers?

No. Page-number stamping runs entirely in your browser using pdf-lib. Your file is loaded into a sandboxed memory buffer, the new text objects are added to each chosen page, and the modified file is delivered as a download. Verify in DevTools Network — no outbound requests are made beyond the page assets themselves. This matters for sensitive documents (contracts, exam papers, internal memos) that need numbers before distribution but should not leave your device.

Can I add Roman numerals (i, ii, iii) for front matter and Arabic for body content?

Yes — run the tool twice. First pass: "Pages to number = 1-4", format = lower-Roman, start = 1, giving i, ii, iii, iv on the first four pages. Second pass: load the output of the first pass, set "Pages to number = 5-", format = numeric, start = 1, giving 1, 2, 3, ... on the rest. This mirrors traditional book-printing conventions where prefatory material (preface, contents) uses Roman and the main body restarts at Arabic 1.