Open Split PDF, keep the mode on By range, and type the pages you want. A single number, a run like 4-9, or a mix such as 1,3,7-9 all work, and they can be combined in one line.
You get a new PDF containing only those pages. The source file is never modified, so there is nothing to undo if you aim wrong.
A signature page out of a twenty page contract. The one chart worth showing from a quarterly report. A single invoice from a scanned batch. These are not really splitting jobs, because you do not want the document divided up. You want a small piece of it, standing on its own.
Extracting a selection
The whole operation is one field. What makes it worth a minute of attention is that the field does more than it looks like it does.
Open the original and count positions
Page numbers here mean position in the file, counting from one. They are not the numbers printed on the page, which is where most mistakes start.
Type the selection in one line
Single pages and ranges can be mixed freely. 1,3,7-9 is a perfectly ordinary thing to type and produces one file with five pages in it.
Download and check the result
The output opens in any reader. Confirm you caught the right pages before the original leaves your mind, because in a week you will not remember what page nine was.
Pulling pages that are not next to each other
This is the capability people most often do not realise they have. Extraction is not limited to a continuous block. Comma-separated selections gather scattered pages into one tidy document.
Take a fifty page lease where the parts that matter are the first page, the rent schedule on page eleven, and the signature block on pages forty-eight and forty-nine. Typing 1,11,48-49 produces a four page file with exactly that, in that order, ready to send to someone who does not need the other forty-six.
- One page on its own
7 - A continuous block
12-20 - Scattered single pages
1,11,48 - Blocks and singles together
1,11,48-49 - Deliberately out of order
48,1,11
9,1,4 produces a three page document with page nine first. Most of the time you want ascending order and type it that way without thinking, but when you need a specific sequence, the range field will give it to you without a separate reordering step.What the extracted pages carry with them
Extraction copies page objects rather than re-rendering them, so the visual result is identical to the source. That much is straightforward. What is less obvious is everything else attached to those pages.
No decode, no re-encode. A 600 DPI scan arrives as a 600 DPI scan.
Extraction reads the source and writes a new file. Nothing is modified in place.
Metadata, annotations and anything embedded in those pages travels with them.
That last column is the one to think about. An extracted page brings its annotations, its form fields, and any metadata attached to the objects it references. If you are extracting a page to share it precisely because the rest of the document is confidential, the page itself may still carry traces of that context.
Where this comes up
Extraction tends to show up at the boundary between a document and a person who should only see part of it.
- Signed contractsExecution pages only
- Long reportsOne chart or appendix
- Scanned batchesOne invoice from many
- Job applicationsA single portfolio piece
- Legal disclosureSpecific numbered exhibits
- Medical recordsOne result, not the file
When the file will not open
Encrypted PDFs cannot be extracted from while they are encrypted, since reading the page objects is exactly what the encryption prevents. Remove the password from a document you own, extract from the unlocked copy, and protect the output again if it needs it.
A damaged file is a different problem. If a PDF opens with errors or shows blank pages, extraction will faithfully copy whatever broken structure it finds. Repair the document first, then extract from the repaired version.
Extract pages in your browser
Type the pages you want and download them as a new PDF. The original stays untouched.
Common questions
Questions that come up when the selection gets complicated.
01Does extracting pages change the original file?
No. Extraction reads the source and writes a new document containing your selection. The original is not modified, moved or re-saved, so there is nothing to undo if you pick the wrong pages. Just run it again with a corrected range.
02Can I extract non-consecutive pages into one file?
Yes, and this is the most useful thing the range field does. A comma-separated list mixing single pages and ranges, such as 1,3,7-9, collects all of them into one output PDF in the order you typed.
03Does extracting reduce quality?
No. Pages are copied as complete objects with their fonts and images intact, not re-rendered or recompressed. The extracted page is byte-for-byte the same content that was in the source.
04Why does my one extracted page have a large file size?
Because size follows content. If that page carries a full-page image or a high resolution scan, it was always a large share of the document's weight. Page count and file size are only loosely related.
05What if the PDF is password-protected?
Remove the password from your copy first, then extract from the unlocked version. Encryption specifically prevents the object-level reading that extraction depends on, so there is no way around this step for a file you have the rights to.
06Do comments and form fields come across?
Usually yes, and that is worth checking. Annotations attached to an extracted page generally travel with it, as do form fields. If the page is leaving your organisation, open the output and confirm nothing internal came along for the ride.