Number PDF Pages Without Uploading (Private, Free) [2026]

The privacy problem with online PDF numbering tools

Search for "add page numbers to PDF online" and you will find dozens of tools that do the job. The problem is that almost all of them require you to upload your file to their server. Your PDF goes across the internet, gets processed on hardware you do not control, and comes back numbered. The service promises to delete it after a period.

For a recipe PDF or a travel itinerary, this is a trivial concern. For the documents that most commonly need page numbers — academic theses, legal briefs, medical reports, financial forecasts, and draft contracts — it is not. These documents may contain unpublished research, privileged communications, protected health information, or commercially sensitive terms. Uploading them, even briefly, creates risk that cannot be unrolled.

I built SammaPix PDF Page Numbers to eliminate this problem. The tool runs the entire numbering process inside your browser using pdf-lib — a mature, open-source JavaScript library for reading and writing PDFs. No server is involved at any stage.

The chain-of-custody problem

When you upload a confidential PDF to an online service, you create a chain-of-custody question that cannot be easily answered: who processed the file, was it cached on a CDN, did it pass through a logging system, could the service's staff access it, and was the deletion actually performed? These are not hypothetical concerns — they are exactly the questions that legal teams, IRB reviewers, and institutional compliance offices raise when they discover staff have used consumer web services to process sensitive documents.

Running the numbering in your own browser means the chain of custody is trivially short: the file is on your device, processed by your browser, and the output is on your device. There is nothing else to audit.

How browser-based PDF numbering works technically

Understanding the technical mechanism explains both why this works and why the results are better than alternatives like printing to PDF from Word (which destroys the original PDF structure). Here is what happens when you click Add Page Numbers:

1. FileReader API loads the PDF into browser memory. The FileReader API is a standard browser API that reads local files without any network access. The PDF bytes are loaded into an ArrayBuffer in memory.
2. pdf-lib parses the PDF structure. The library traverses the PDF's cross-reference table, page tree, and object dictionary. It builds an in-memory representation of every page, including each page's dimensions (MediaBox) and its content streams.
3. For each page, a new content stream is appended. pdf-lib adds a content stream that draws the page number as a text object at the calculated position. The position is computed from each page's MediaBox dimensions, so numbers are correctly placed relative to the actual page size regardless of whether the document mixes page sizes.
4. The text is written as real vector text, not an image. The content stream uses PDF text operators (BT ... ET) with font references and position coordinates. This is the same way the original body text of the document was written into the PDF. The numbers are real text — selectable, searchable, infinitely scalable without pixelation.
5. The modified PDF is serialized and downloaded. pdf-lib produces the output PDF as a Uint8Array. A Blob is created and downloaded via a blob: URL. No network request is made at any point during this step.

The critical distinction from rasterization tools: pdf-lib never converts the PDF pages to images. Everything operates at the PDF data structure level. The original content — fonts, text streams, images, vector graphics, hyperlinks, form fields — is preserved byte-for-byte. The only addition is the new page-number content stream on each page.

Custom start number: the feature professionals actually need

Most page-numbering tools only offer one option: start at 1. This is fine for standalone documents. It is completely wrong for documents that are part of a larger sequence. The custom start number field is the feature that makes this tool genuinely useful for professional workflows.

Thesis and dissertation assembly workflow

Academic theses are commonly written in separate chapter documents and assembled into a single PDF only at the end. If you are working in this way and need to share an individual chapter for advisor review — with correct page numbers — you need to set the start number to whatever page that chapter begins on in the full document.

Example: your full thesis will run to 214 pages. Chapter 3 covers pages 47 to 89. Upload chapter 3 as a standalone PDF, set start number to 47, and the numbering on page 1 of that PDF will read 47. When assembled with the other chapters, the numbers will match perfectly.

Legal brief and court filing workflow

Legal filings frequently consist of a main brief plus multiple appendices, each submitted as separate PDFs but paginated continuously. The main brief runs pages 1 to 45. Appendix A runs pages 46 to 58. Appendix B runs pages 59 to 67. Each appendix PDF is numbered independently with its correct start number, so any page reference in the brief or the appendix is unambiguous regardless of which section the reader is looking at.

Corporate report assembly

Annual reports, audit findings, and due diligence packages are often assembled from sections contributed by different teams. Each team exports their section as a PDF. Before final assembly, each section PDF can be numbered with the correct start number so that the final merged document has consistent pagination matching the table of contents — which is prepared separately with knowledge of how many pages each section runs.

Choosing formats and positions for academic, legal, and business documents

Different document types have different conventions. Here is a practical guide to choosing the right combination:

Academic documents (theses, dissertations, journal submissions)

The majority of academic style guides (APA 7th edition, MLA 9th edition, Chicago 17th edition) place page numbers in the header at the top right or in the footer at the bottom center. APA specifically requires the running head and page number in the top right for manuscripts. Chicago-style papers typically use bottom-center. Check your institution's specific thesis formatting requirements — many universities publish a style guide that specifies exact position and format.

Format: most academic submissions use the plain number format. The Page N of N format is increasingly common for theses where the committee wants to verify completeness.

Legal documents (briefs, filings, contracts)

Court rules vary by jurisdiction. Most US federal courts require page numbers at the bottom center or bottom right. The Page N of N format is explicitly required in some jurisdictions for certain filing types. Review the relevant court's local rules or the style requirements of the contracting party. For contracts, bottom right is a common professional convention that leaves the document's header area clean for the party's letterhead.

Business reports and proposals

Business documents often have branded headers at the top. Place page numbers at the bottom — center or right — to avoid overlap. The current/total format (e.g., 3 / 12) is popular for proposals and pitch decks because it signals to the reader how much remains without requiring them to scroll to the end.

How to number PDF pages without uploading, step by step

The full process takes under a minute:

1. Check for modification restrictions. If the PDF was generated by a legal service, form platform, or e-signature tool, it may have owner-password restrictions that prevent modification. Try the tool and it will let you know. If restrictions are present, remove them first using PDF Unlock — also in-browser, no upload.
2. Go to sammapix.com/tools/pdf-page-numbers in Chrome, Safari, Firefox, or Edge. No account required.
3. Drop your PDF onto the dropzone. The file loads into browser memory. Nothing is transmitted.
4. Set the start number. The default is 1. Change it if this PDF is part of a larger sequence.
5. Choose position and format. Refer to the guidance above for academic, legal, and business conventions.
6. Adjust font size if needed. The default works for standard A4/letter documents.
7. Click Add Page Numbers. pdf-lib processes the file in memory. Seconds for most documents.
8. Download and verify. Open the downloaded PDF in any viewer. Select a page number — it highlights and copies as text, confirming it is a vector text object, not an image.

Why printing to PDF from Word is not the right approach

A common workaround is to open the PDF in Word (which converts it), add page numbers in Word, then print back to PDF. This approach has significant problems:

• Layout destruction. Word's PDF import is a conversion, not a lossless read. Tables shift, fonts substitute, images reflow, and text spacing changes. For complex documents — multi-column layouts, footnotes, legal formatting — the result is often unusable.
• Hyperlink breakage. Internal PDF links (table of contents links, cross-references) frequently break during the Word import and re-export cycle.
• Font substitution. If the PDF uses embedded fonts that Word does not have, Word substitutes them with fallback fonts, changing the visual appearance and potentially the line breaks of the document.
• Color profile changes. PDFs using CMYK or ICC color profiles can have their colors shifted when round-tripped through Word's RGB-centric workflow.
• Metadata loss. Custom metadata fields, document properties, and PDF/A compliance settings are typically stripped during the Word round-trip.

pdf-lib adds page numbers without any of these problems because it operates on the PDF at the data structure level. It reads the existing objects and appends new ones. No conversion, no round-trip, no layout engine involved.

Browser-based vs upload-based PDF numbering: privacy-first comparison

Here is an honest comparison across dimensions that matter for privacy-sensitive documents:

How to verify no upload happens

For a privacy claim this important, you should not take my word for it. Here is how to verify it in under two minutes using your browser's built-in tools:

1. Open DevTools. Press F12 (Windows/Linux) or Command Option I (Mac). On Safari, enable the Develop menu in Settings → Advanced first.
2. Go to the Network tab and clear it. Click the Network panel. Use the clear button to remove any existing requests. This gives you a clean log to watch.
3. Drop your PDF and run the tool. Watch the Network panel as you drop the PDF, configure the settings, click Add Page Numbers, and click Download.
4. Observe the results. You will see requests for JavaScript, CSS, and font assets when the page first loads. During numbering and download, you will see zero outgoing requests. No request carries your PDF bytes anywhere.

This verification method is the same one used by security researchers and compliance auditors to confirm that a tool genuinely processes data locally. It is conclusive: if no bytes of your file appear in outgoing requests, the file never left your device.

Other browser-based PDF tools for document workflow

SammaPix offers a complete suite of browser-based PDF tools. All run in your browser with no upload. Here is how they fit into a typical document preparation workflow:

• PDF Page Numbers: add page numbers — six positions, three formats, custom start number. Vector text. No upload.
• PDF Unlock: remove modification and copy restrictions before numbering. See the full guide at Unlock a PDF online without uploading it.
• PDF Merge: combine sections before adding continuous page numbers to the merged whole.
• PDF Split: extract the relevant pages before numbering a section, if working with a larger source document.
• PDF Compress: reduce the file size of the final numbered document before sending. See Compress a PDF online without uploading it.
• PDF Rotate: correct page orientation before numbering to ensure numbers appear on the right edge of each page.