Splitting sounds like one operation. It is really three, and picking the wrong one is how you end up with 400 files named page-1 through page-400.
Which kind of split do you actually need?
The three modes produce very different results from the same file, and most tools present them as one button.
| You want | Mode | Output from a 40-page file |
|---|---|---|
| Page 12 on its own | Extract | 1 file |
| Pages 5–9 as one document | Extract range | 1 file |
| Three chapters as three files | Split at boundaries | 3 files |
| Every page separately | Burst | 40 files |
| Everything except page 3 | Extract range twice, then merge | 1 file |
That last row is worth noting because it is the one people expect to be a single operation. Deleting a page is not usually offered as such — you extract what you want to keep, on either side of the gap, and merge the two halves back together.
How to write a page range that does what you mean
Range syntax is near-universal across PDF software, and worth knowing because it saves a great deal of clicking.
7— just page seven.5-9— pages five to nine inclusive. The last page is always included, which trips people up when they write1-10expecting ten pages after page one.1,4,9— three separate pages, in one file, in that order.3-— page three to the end, in most tools.-6— the start through page six.
Two things to watch. Pages are numbered by position in the file, not by what is printed on the page — a report with roman-numeral front matter will have its "page 1" somewhere around file position 9. And the order you list ranges in is usually the order you get, so 9,5 gives you page nine followed by page five.
Splitting a scanned document
A scan behaves differently from a digital PDF in one way that matters here: there is no text layer, so nothing in the file tells you where one section ends and the next begins. You are working from thumbnails.
That makes long scans genuinely tedious, and there are two habits that help. Scan in batches that match the divisions you will want — if a bundle has five documents in it, scan five files rather than one and merge later if needed. And if you already have the single large scan, work from a page-thumbnail view rather than scrolling, so you can spot section starts by their shape.
If you need to find text inside a scan to locate a boundary, run the pages through Image to Text first — OCR will tell you which page the heading is on even though the PDF itself cannot.
What survives a split?
Extraction copies page objects across, so the pages themselves are untouched — same resolution, same fonts, same vector content. No quality is lost. What can go missing is everything that lived at document level rather than page level.
- Bookmarks pointing at pages outside the extracted range break, and many tools drop the outline entirely.
- Internal links to pages you did not extract have nowhere to go.
- Form fields usually survive on the page they sit on, but any calculation referencing another page stops working.
- Digital signatures are invalidated — the signature covered the whole original file.
- Document metadata such as title and author is often not carried over, which matters if the split file will be shared.
Page content itself is safe. If someone tells you splitting degraded their pages, what actually happened is that the tool re-rendered rather than extracted, which is worth checking by comparing file sizes: an extracted range should be roughly proportional to its share of the original.
Splitting to get under a size limit
This is one of the most common reasons to split, and the arithmetic is more forgiving than people assume.
A 60 MB, 300-page manual split into three does not give you three 20 MB files unless the pages are evenly weighted, and they rarely are — the section with all the photographs carries most of the size. Split on content weight rather than page count and you will get closer.
Better still, compress first and see whether you need to split at all. A scanned document at 150 DPI and quality 0.75 typically loses 60–80% of its size with nothing visible gone, which often removes the need entirely. The reasoning is in how to compress a PDF without losing quality, and the email-specific thresholds are in how to compress a PDF for email.
Frequently asked questions
How do I extract one page from a PDF?
Use a split tool in extract mode and give it the single page number. You get a new one-page file and the original is untouched. Page numbers refer to position in the file, not to what is printed on the page, so front matter shifts everything.
How do I delete a page from a PDF?
Most tools do not offer deletion directly. Extract the pages before the unwanted one and the pages after it as two files, then merge them back together. For a single page near the start or end it is faster to extract just the range you are keeping.
Does splitting reduce quality?
No. Extraction copies page objects across unchanged — same images, same fonts, same vectors. If output pages look worse, the tool re-rendered rather than extracted them. Compare file sizes: an extracted range should be roughly proportional to its share of the original.
Will splitting keep my bookmarks?
Usually not. Bookmarks are document-level and any pointing outside the extracted range are broken, so most tools drop the outline entirely. If the bookmarks matter, rebuild them in a PDF editor after splitting.
Can I split a password-protected PDF?
Not while it is encrypted — the tool cannot read the page contents. Remove the protection in a PDF reader using the password first, split, then re-apply protection to the outputs if they still need it.
What is the fastest way to split a 400-page scan?
Not bursting it into 400 files. Work from a thumbnail view to find your section boundaries, then extract each section as a range. If you need to search the text to find boundaries, run OCR first so you can locate headings by page number.