Split, extract or remove: choosing the right way to cut up a PDF

"Split this PDF" usually means one of three different things, and picking the wrong one creates extra work.

The three operations

Split takes one document and produces many. The typical use is one file per page, or one file per fixed-size chunk. You end up with a folder of outputs.

Extract takes one document and produces one smaller document containing only the pages you named. Everything else is discarded.

Remove takes one document and produces one document with the named pages deleted. It is the inverse of extract, and the right choice when you want most of the file.

Choosing between extract and remove is purely about which list is shorter. Keeping 4 pages out of 90? Extract. Dropping 4 pages out of 90? Remove. Same result, a tenth of the typing and far fewer mistakes.

Page range notation

Both Extract Pages and Remove Pages accept the standard comma-and-dash notation: 1,3,5-9,12. Two habits prevent most errors:

• Count from the printed page or the viewer's page number? They frequently differ, because covers and front matter may be unnumbered. The tools use the physical page index — the first sheet is page 1 regardless of what is printed on it. Check in the viewer before typing.
• Ranges are inclusive on both ends. 5-9 gives you five pages, not four.

When to split into many files

Splitting one file per page is genuinely useful in a few situations:

• Archiving documents where each page is an independent record — receipts, certificates, scanned forms.
• Feeding a downstream system that accepts single-page files only.
• Distributing individual pages to different people without exposing the rest.

It is a poor choice for chapters of a report: you get document-1.pdf through document-260.pdf and no way to tell which is which. For that, extract each chapter's range separately and name the outputs yourself.

Naming the outputs

Split tools produce sequentially numbered files, and the numbering is the only thing distinguishing them. Two practical points:

• Rename immediately, while you still remember what the ranges were. A folder of report-page-17.pdf files is unhelpful a week later.
• Use zero-padded numbers if you will re-merge later, so alphabetical sorting matches page order.

Splitting a scanned document

Scans behave the same way structurally, but two extra things are worth knowing.

Each page carries its own image, so the split files sum to roughly the original size. Splitting is not a way to make a large scan smaller — for that see reducing PDF size.

If the scan has been through OCR, the recognised text layer is part of the page and travels with it. Split pages remain searchable. If you split first and OCR afterwards, you pay the recognition cost per file, which is slower.

Reordering rather than cutting

Sometimes the real problem is sequence, not content: a scanner fed pages in the wrong order, or an appendix belongs at the front. Organize PDF shows page thumbnails and lets you rearrange and delete visually, which is far less error-prone than working out ranges by number. Use it whenever you would otherwise be extracting several ranges and merging them back together.

A sensible sequence

1. Open the document and note the physical page numbers you care about.
2. Decide: many files (split), one subset (extract), or original-minus-a-few (remove).
3. Run the operation and open the output before deleting the source.
4. If the pieces need reassembling in a new order, merge them rather than trying to do everything in one pass.

All four tools run in your browser, so cutting up a confidential document does not involve uploading it.