How to Add Page Numbers to a PDF Without Paying for Acrobat

"We can't file this. The pages aren't numbered."

Court filings reject unnumbered documents. California's rule for electronic filing spells this out directly: a document filed electronically "must begin with the first page or cover page as page 1 and thereafter be paginated consecutively using only arabic numerals" (Cal. R. Ct. 8.74), with the same cover-page exception most people actually want built in — the rule lets a cover page's own number be suppressed. Universities publish comparably specific pagination rules for theses and dissertations. Contracts reference "page 14, section 3" and fall apart when nobody can agree which page is 14. Page numbers are boring infrastructure until their absence blocks something important.

Quick answer: Open Add Page Numbers, drop in your file, pick a position and format, set a start number, and download. The numbers are stamped into the page content itself, so they print and travel with the file — free, in your browser, no upload.

Adobe charges for the privilege of adding them. As of September 2026, Acrobat Standard starts at $14.99 a month billed annually ($24.99 month-to-month), and Acrobat Pro starts at $19.99 a month annually ($29.99 month-to-month). For stamping digits in a corner, that's a steep toll — and subscription pricing moves, so treat the exact figure as a snapshot rather than a promise.

Viewer Numbers Are Not Page Numbers

Here is the confusion that derails people: your PDF reader already shows page numbers. The toolbar says "page 12 of 40." So the document is numbered, right?

No. A page number, in the sense a court or a citation cares about, is text drawn directly into a page's own content stream — permanent ink that travels with the file wherever it goes. The toolbar counter is something else entirely: the viewer's own arithmetic, the nth page of the file, recalculated fresh every time you open it. It has four problems:

1. It doesn't print. The counter lives in the toolbar, not on the page. Print the document and the numbers vanish.
2. It ignores front matter. A thesis with 8 pages of Roman-numeral front matter shows "page 9" where the printed page 1 begins — the Roman-then-Arabic split is a long-running convention that graduate schools still document today, including Clemson's Graduate School, even at schools where it's since become optional rather than mandatory. Legal citations to printed numbers and viewer numbers diverge either way.
3. It can't skip pages. Cover sheets and title pages shouldn't carry numbers. The viewer counts them anyway.
4. Nobody else sees your position. "Look at page 12" means the 12th page of your viewer state, which may not match anyone else's — different reader, different zoom, sometimes a different result.

Real page numbers are content stamped onto the page: they print, they export, they survive being sent to someone with a different reader.

The Decisions That Actually Matter

Position

Bottom center is the default for a reason — it never collides with headers, footers, or marginalia. Bottom-right suits business documents. Top-right is common in legal filings. OxygenPDF's Add Page Numbers tool exposes all nine reasonable placements as a 3×3 grid — top, middle, or bottom, crossed with left, center, or right — instead of picking two or three "common" options and making you live with the rest. Whatever you pick, check every page: a full-page diagram or a landscape table can swallow a corner-placed number whole.

Format

Plain digits work almost everywhere. "Page X of Y" helps readers gauge document length but requires knowing the total at stamping time, so lightweight tools tend to skip it — OxygenPDF's does too, for the same reason. What it offers instead is six formats: decimal (1, 2, 3), a padded variant that zero-fills single digits only (01 through 09, then plain digits from 10 on), Roman numerals in upper or lower case (I, II, III / i, ii, iii), and letters in upper or lower case (A, B, C…Z, then AA, BB — the same base-26 scheme spreadsheet columns use, useful for lettering appendices or exhibits instead of numbering them).

Roman numerals for front matter plus Arabic for the body is the traditional thesis convention, and the most commonly botched setup, because it needs two numbering passes with different formats and different starting values applied to different page ranges. A single-pass tool can only do one format across one file — see "Can I Use Roman Numerals for the Front Matter and Arabic for the Body?" below for how to get both into one finished document anyway. While the thesis is open, making its citations clickable saves your examiner a lot of scrolling.

Starting number and range

Numbering rarely starts at 1 on the first page of the file. Cover pages, title sheets, and tables of contents stay unnumbered, and the body starts at 1. OxygenPDF's tool gives you two controls for this: a start number (1 to 9999) that sets what value the first numbered page shows, and a "skip first page" toggle that exempts exactly one page — typically the cover — from getting a number at all, without shifting the count underneath it. Turn both on with the start number at 1, and page 2 of the file becomes "1." It's the same outcome Rule 8.74 describes when it says a cover page's number "may be suppressed and need not appear."

What a "skip first page" toggle won't do is number an arbitrary range in the middle of a file — pages 5 through 40 only, say, or a multi-page front matter followed by a differently-formatted body. A tool that can only skip one page and number everything else from a single start value forces you toward the split-number-merge workflow described further down. That's more steps, but it's no less reliable, and it's exactly what a multi-page front matter needs regardless of which tool you use.

Typography

The tool draws every number in one clean, black, sans-serif face, sized from 6pt to 72pt, with a margin of 0 to 200pt from the page edge — there's no font picker and no color swatch in the interactive version. That's a reasonable default, not an oversight: black sans-serif numbers read cleanly against nearly any body typeface, serif or sans, without competing for attention. Size it to be findable but invisible: 9-11pt is where most house styles land. If a document specifically calls for a different color or typeface on the running number — some in-house style guides ask for it to match a header's charcoal gray — that's outside what a dedicated numbering tool is built for; a general-purpose PDF editor gives you that level of control, at the cost of everything else this one keeps simple.

Why Not Upload It

Documents needing page numbers skew formal and confidential: court filings, merger agreements, medical trial protocols, unpublished theses. These are exactly the files that shouldn't transit through a free tool's server for a cosmetic operation.

Numbering is text stamping — drawing small strings at coordinates on each page. It needs no server, no AI, no account. The browser is fully capable, and nothing about the file leaves your device to get it done. One thing numbering won't do is hide anything: if a page carries information that shouldn't leave your hands at all, redact it properly before the numbers go on.

Numbering with OxygenPDF

1. Open the Add Page Numbers tool
2. Drop your file in
3. Pick a position from the 3×3 grid, a number format, font size, and margin
4. Set a starting number, and toggle "skip first page" on if you have a cover sheet
5. Download the numbered file

Everything runs locally. The output downloads as <original-name>-numbered.pdf, so re-running the tool on the same file never silently overwrites your source. Check the result by scrolling the full document — especially landscape pages and full-bleed figures where numbers hide.

Drop in more than one PDF at once and the tool switches to batch mode automatically: configure position, format, and starting number once, and it applies that same configuration to every file in the batch, downloading each as its own numbered PDF when it finishes. One thing batch mode doesn't do is carry a number forward between files — each file starts over at your chosen start number rather than continuing from wherever the previous file left off. For a multi-file production that needs one continuous, non-restarting count across every document, that's a job for Bates numbering rather than page numbering (more on that distinction below).

If your file is password-protected

Unlock it at the password prompt and numbering proceeds normally — but it's worth knowing what happens underneath. A tool that stamps numbers onto a page needs a genuinely decrypted file to work with, so a protected PDF gets decrypted by rendering each page and rebuilding an unencrypted copy before the numbers go on. If the document was scanned or image-based to begin with, nothing changes for you. If it had a real, selectable text layer, that layer doesn't survive the round trip: the output is fully readable and printable, numbers included, but the underlying text is no longer selectable or searchable the way the original was. If you need both a numbered and a text-searchable copy of a protected file, keep the original alongside the numbered output rather than treating the stamped copy as a full replacement.

Can I Use Roman Numerals for the Front Matter and Arabic for the Body?

Not in a single pass through a tool that numbers one file with one format at a time. The workaround is four steps instead of one, but each step uses a tool built for exactly that job:

1. Split the document into a front-matter file and a body file (splitting PDF pages walks through the options).
2. Run Add Page Numbers on the front-matter file with the Roman format, skipping the title page if it shouldn't carry a visible number.
3. Run it again on the body file with the decimal format and a start number of 1.
4. Merge the two numbered files back into one document, front matter first — the same ordering logic that applies when merging PDF files generally.

The result is indistinguishable from what a paid editor's "section-based" numbering produces. A PDF file doesn't record which application stamped a number or how many passes it took — the numbers are just text on the page either way.

Edge Cases Worth Knowing About

Scanned or image-only PDFs. Numbering doesn't need existing text to work. It draws a number on top of whatever is already on the page, image or otherwise, so a scanned contract numbers exactly as cleanly as a native one.

Mixed page orientations. Position is calculated from each page's own width and height, so a landscape page in the middle of a portrait document still gets a correctly placed number rather than one computed from the wrong dimensions. It can still land inside artwork that runs to the edge of a landscape page, though — check those pages specifically after processing.

Running the tool twice on the same file. Nothing stops you from numbering an already-numbered file a second time, and the result is two numbers stacked on the same spot. Keep an unnumbered original (see Practical Tips) rather than relying on catching the overlap visually before you send the file anywhere.

Double-sided, bound printing. Some house styles alternate the number's position between the inside and outside margin on facing pages, so the numbers stay clear of the spine. A single position setting applies to every page here; there's no separate odd/even placement. A bound document that needs mirrored margins is a job for a general PDF editor, not a numbering-only tool.

Continuing numbering from a previous volume. The start-number field accepts values from 1 to 9999, so you can pick up where an earlier binder or exhibit volume left off — set it to 9847 and stamping continues from there. Once processing starts, the running count itself has no ceiling; the 9999 limit is only on what you can set as the very first value. For a genuinely large, multi-file production where you want a unique identifier per page rather than a number that restarts in every file, that's what Bates numbering is for — see below.

Practical Tips

Number last. Add numbers after merging, splitting, and reordering are finished. Every structural change reshuffles which content sits on which number, and renumbering a finished document beats tracking a moving target.

Keep an unnumbered copy. If the filing gets rejected for an unrelated reason and you revise, you'll restamp from the clean original instead of layering numbers over numbers.

Page Numbers or Bates Numbers — Which Do You Need?

A page number restarts at 1 within a single document and describes position inside that one file. Bates numbering is different: a unique, sequential identifier — typically a prefix plus zero-padded digits, like ACME-000142 — applied across an entire set of documents in a legal production, so any single page can be found and cited without ambiguity no matter which of the underlying files it came from. It's a convention rather than a codified format (there's no single mandated prefix style or digit count), but it's the one discovery production has run on for well over a century, originally with hand-cranked stamping machines and now digitally.

If you're preparing a filing on its own, plain page numbers are enough. If you're producing documents for discovery, or numbering exhibits that need to survive being cited out of context, use OxygenPDF's Bates Numbering tool instead — it adds the prefix, suffix, and digit-padding controls that page numbering intentionally leaves out.

Add page numbers here. It stays in your browser.