How to extract pages from a PDF
You need one month out of a year of bank statements, or the signed page of a forty-page contract. That is extraction — and it is not the same thing as splitting a document, or as deleting pages from it. Mixing the three up is the reason people end up with a folder full of files they did not want. Here is the difference, and then the two-minute job itself.
Split PDF →Extract, split, delete: three different outcomes
All three start with the same action — choose pages — but they leave you with different things. Read the table once and the choice stops being confusing.
| Job | What you select | What you get | Use it when |
|---|---|---|---|
| Extract | The pages you want | A new file with just those pages; the original is untouched | You need to send a subset and keep the whole |
| Split | Each range in turn | Several files that together cover every page | A long report has to become sections |
| Delete | The pages you do not want | The original document minus those pages | Blank backs, duplicates, a page that must not be shared |
Extracting and splitting are done from the same screen with Split PDF, because both come down to “choose pages, save them as a file”. Select pages 12-15 and save once and you have extracted them; select 1-20, save, then 21-40, save, then 41-60 and you have split the document. If what you want is the original minus a few pages, that is the third job — delete PDF pages instead and keep everything else in place.
How to pull the pages out
The whole operation takes about as long as reading this section.
- Load the PDF — drag it onto the page or click Choose file. Every page appears as a numbered thumbnail.
- Tick the checkbox on each page you want to keep. The counter at the top shows how many are selected, and Select all and Clear selection help in long documents.
- Click Selected to download exactly those pages as a new PDF.
- Change the selection and repeat if you need a second extract. The original file on your disk is never modified, so you can take several different sets from the same source.
There is no minimum and no upper limit: one page is fine, three hundred is fine. The document is opened and rebuilt inside your browser and never sent anywhere, which is the practical reason to do this with statements, medical records and ID packets here rather than on a service that uploads them first.
Printed page numbers lie — work from the thumbnails
This is the single most common way an extraction goes wrong. The number printed on a page and the page's real position in the file disagree far more often than people expect: an unnumbered cover, a title page, roman numerals in the front matter, a scanned insert or an inserted map shifts everything by one or two.
The tiles here are numbered by physical position, which is what the selection uses, and each tile shows you the actual page. Glancing at the picture before you save removes off-by-one errors entirely — and off-by-one is exactly the error that sends the wrong page of a contract to the wrong person.
Two checks worth making before you download. Count: if you meant to take four pages, four tiles should be ticked. And look at the pages either side of your selection, because the classic mistake is grabbing the blank back of the signature page instead of the signature page itself.
What you actually get back
The saved file contains those pages exactly as they were. Nothing is re-rendered or re-compressed, so text remains selectable and searchable, scanned images keep their resolution, and the extracted file weighs roughly what those pages weighed inside the original.
That means extraction is also a legitimate way to make a file smaller when a form has a size limit — sending three relevant pages instead of sixty solves the problem without touching quality at all. It is worth trying before compression for that reason.
Your original is only ever read. Nothing is committed until you press download, so if the result looks wrong, reload the file and start again — there is no damage to undo.
What people actually pull out
Nearly every request comes down to sending less than the whole document, whether for privacy, for size, or out of courtesy to the reader:
- One month from a year-long bank statement, for a landlord or a loan application, without handing over eleven other months of spending.
- The signed page of a contract, to email back on its own rather than returning forty pages the other party already has.
- A single chapter of a textbook or manual, small enough to read comfortably on a phone.
- One invoice out of a batch export, to attach to a specific expense claim.
- The two pages of a scanned application a clerk asked you to resend, when the rest of the packet contains ID scans you would rather not circulate again.
- A handful of slides from a long deck, so a colleague opens the file on the part that concerns them.
A few things pair naturally with extraction. If the pages came out sideways, straighten them with Rotate PDF before saving. If you need the parts back in one document later, or in a different sequence, Merge PDF puts them together again. And if the extract is still too heavy to email — which happens with scans — Compress PDF is the last step, not the first.
Common questions
What is the difference between splitting and extracting?
Splitting cuts one document into parts so that every page ends up somewhere — a 60-page report becomes three 20-page sections. Extracting lifts a subset out, such as pages 12 to 15, while the rest of the document carries on existing untouched. The same screen does both.
Can I extract a single page?
Yes. Tick one tile and click Selected, and that page is saved as a one-page PDF. There is no minimum and no upper limit.
Does the original file change?
Never. The file on your disk is only read. What you download is a new document containing the selected pages, so you can extract several different sets from the same source without damaging it.
Will the extracted pages lose quality?
No. Pages are copied across without being converted or re-encoded, so a text page stays searchable and sharp at any zoom, and a scan keeps the resolution it was scanned at.
Is there a limit on file size or page count?
Only what your device can hold in memory. There is no server-side cap of 20 MB or 100 pages, because there is no server. A heavy scanned document may take a few seconds to display all its thumbnails, but it will open.
How do I remove pages instead of extracting them?
Use the delete-pages tool, which keeps everything you do not tick. Extraction gives you a new file containing only the selected pages; deletion gives you the original document with those pages taken out.