← All posts

Split a PDF into single pages, one file per page

Something on the other end wants one page at a time. A council portal that takes a single PDF per field, a print shop that files by sheet, a records system that treats a document as one page, a colleague who asked you to "send them separately". You have a 40-page file and a deadline.

Every tool you have tried asks where to split. You do not want a where. You want all of it, every page, each in its own file, and you would rather not click 40 times.

That takes about twenty seconds. The part that actually goes wrong is what the files end up being called, and it goes wrong quietly — usually after you have uploaded them in the wrong order.

Why each page has to become a whole document

A PDF page is not a self-contained thing. The page object describes what goes where; the fonts, the images, the colour profiles and the logo in the letterhead live in a shared pool that the whole document draws on. Page 12 is a set of instructions pointing at objects that pages 1 to 40 also point at.

So "one file per page" cannot mean cutting the file into forty slices. Each page has to be rebuilt as a new document, with a copy of every object it references written into it. That is why the operation is safe — nothing is re-rendered, so nothing loses quality — and it is why two things that surprise people happen:

If you actually meant half the document rather than every page, that query has two different answers and they need different tools.

Doing it

  1. Check the page count first if you do not already know it — our PDF inspector reads it out of the file, and you will want the number in a minute to check the result.
  2. Open Split PDF by pages and upload the document.
  3. Ignore the "split after these pages" box. That is the other way of using this tool. Leave it alone.
  4. Set "or split every N pages" to 1. One page per file is just the smallest possible run. The number here wins over the page numbers above, so you do not need to clear anything.
  5. Download the zip and unpack it. Inside are your single-page PDFs, in page order.
  6. Count them. Forty pages in, forty files out. If the numbers do not match, something was wrong with the upload, not with the split.

The names, and why page 10 files itself between 1 and 2

The parts are named after your document with a number on the end: report-1.pdf, report-2.pdf, through to report-40.pdf. The number is the part's position, counting from 1.

There is no zero padding, and that is the detail worth knowing. Windows Explorer and macOS Finder both sort file names naturally — they spot the digits and read them as a number — so on your own desktop the order looks perfect. Almost everything else sorts byte by byte, character by character, where 1 comes before 2 and so report-10.pdf sorts immediately after report-1.pdf:

report-1.pdf
report-10.pdf
report-11.pdf
report-12.pdf
report-2.pdf

That is the true order in the zip listing, in a shell, in most upload dialogs and in most systems that ingest a folder.

Padding fixes it permanently. One pass, before you touch anything else:

Once every number is the same width, byte order and human order are the same thing everywhere.

What this will not do

It will not name the files after what is on them. Nothing reads page 7 and decides it is "Schedule 2". The number is the position and that is all the information there is. If your document has an outline, Split PDF by bookmarks is the better tool: it cuts at the bookmarks and names each part after the bookmark it starts at, so you get Schedule 2 - Charges.pdf instead of report-4.pdf. Our Rename PDF tool takes one file at a time and is for giving a single document a proper name and title, not for renaming a folder — the commands above are the right tool for that.

It will not keep bookmarks or form fields. For the reason in the mechanism above. If the document is a filled form and you need the answers to survive as part of the page, flatten it before you split.

It will not take one page. Well, it will, along with all the others. If you only want pages 4 to 9 as a document, Extract PDF pages is one step instead of a split and a hunt through a zip. And if you split too far and want a few of the parts back together, Merge PDF is the way back.

It has limits. 50 MB per upload and 1,500 pages. A one-page document is refused rather than handed back to you inside a zip containing itself.

When the command line is genuinely better

If this is a weekly job, or the file is a 3,000-page scan, do it locally. Both of these pad the numbers for you, which removes the whole sorting problem:

Worth saying, because it is the other thing people search for: mutool has no burst mode. It will happily convert, merge and clean a PDF, but there is no one-line "split into single pages" in it — which is why that question keeps being asked and never quite answered.

Before you split, and after

Two things that save a second pass.

If the document came off a duplex scanner, half of your forty files will be the blank backs of single-sided sheets. Drop them first — finding them is the actual problem, because a scanned blank page is never actually blank — and you will produce twenty files rather than forty and a deletion job.

If the pages are not in the order the receiving system expects, fix the order in the document with Organize PDF pages before splitting, not in the file names afterwards. Writing 3,1,2 once beats renaming forty files twice.

Then check the count against the page count, open the first file and the last file, and send them.

Tools this post uses: