Split a PDF on your computer without uploading the packet

One scan is rarely one job. A 40-page pack from a scanner is often a cover letter, an ID page, a bank statement, and a form you must not send together. Portals still want “PDF only, this many megabytes, this page range.” The usual fix is a website that accepts the whole upload, slices it, and emails a link. That is a poor fit when the pack includes other people’s data sitting next to the two pages you actually need.

Split PDF on Tool Hub runs in your browser tab. You pick the file on your computer. You choose pages or ranges. You download the pieces. We do not take a copy to split on a server.

When split is the right tool

Use this page when the file is already a PDF and you need a subset. Typical jobs: extract the signature page from a contract; send only the invoice pages from a monthly pack; pull the photo page from a scanned passport PDF without mailing the whole booklet scan.

This is not merge. If you have five files and need one packet, use Merge PDF. This is not OCR. Split will not make a photograph of a whiteboard into searchable text. This is not redaction. Cutting a page out is not the same as blacking a number on a page you still send. If a number on the page you keep must disappear, that is a different product, and it is not live on this host yet — do not send that PDF to a random “redact online” upload while you wait.

I use split when the scanner dumped everything into one file because that was the default. I do not use it to “hide” pages by hoping the recipient will not look. If a page is in the file you upload to a government portal, assume they can see it. Split first, then send only what the form asked for.

What “in the browser” means for this job

Client-side means the page sends code to your tab. Your browser reads the PDF with the File API. It copies the chosen pages into a new PDF with a library that runs in the tab (pdf-lib on this hub). When you click download, the new file is saved through the browser download manager. We never had a copy to delete, because we never received one.

You can test that. Load the page, disconnect the network, pick a throwaway PDF, extract a page. If it still downloads, the bytes did not need our server for that step. Encrypted PDFs and odd scanner exports can still fail locally. That is a file problem, not a hidden upload.

How I split a file here

1. Open https://tools.sw-commercial.com/split-pdf.
2. Pick the PDF from your computer. Do not email it to us.
3. Choose individual pages or a range (for example 3–5). Count from the file you opened, not from a printed sticker on a folder.
4. Check the preview if the tool shows one. Scanner PDFs sometimes have a blank first page. Including it is how people mail “empty” attachments.
5. Download the result. Name it for the recipient (statement-june.pdf), not split-output.pdf, so you do not attach the wrong piece later.

You do not create an account first. Free split does not watermark the pages.

Failure cases I hit while using it

Page numbers in the PDF viewer are not always the same as printed footer numbers. A “page 12 of 40” in the footer can be file page 13 if there is a cover. I look at the thumbnail, not the footer, before I extract.

Password-protected PDFs may not open in the tab. Unlock them in a tool you already trust (your own reader), save a copy, then split. Do not paste the password into a random website.

Very large scans (hundreds of pages, photo-heavy) can be slow in the browser. That is your machine decoding pages. Split the job: extract the range you need rather than asking the tab to materialize 200 page images. If the tab crashes, try a smaller range.

A “PDF” that is actually a folder of images glued by a phone app can split into pages that are still photos. That is expected. Split does not OCR them.

When not to use this page

Do not use split as a privacy guarantee. The pages you keep still contain whatever was on them. GPS in an embedded JPEG, comments, and form fields can survive a page copy. If you must strip photo location, run the image tools on exported pages, or print-to-PDF from a viewer after you have checked properties. That is extra work. It is honest work.

Do not use an upload splitter for a pack that includes other people’s IDs “just this once.” The portal’s size cap is not consent to store the rest of the family on a converter.

Do not split and then merge the same pages back as a way to “clean” a file. That cargo-cults privacy. If you need a clean export, start from the pages you intend to send.

Related jobs on the same hub

Merge PDF is the sibling. JPG to PDF is for photos that are not PDFs yet. SW Toolbox can split from a PDF link in Chrome, still on the device.

This catalog only publishes how-tos for destinations that return HTTP 200. Split PDF on Tool Hub does. Other family names (redact, OCR, planners) are launching soon and are not linked from this guide.