You've got a 140-page contract and the other side only needs the schedule on pages 88 to 93. Or a full bank statement when the landlord asked for the page showing your salary. Sending the whole thing is lazy at best and an overshare at worst. You want a new PDF with just those pages.
Extract pages does exactly that. You pick the pages, it builds a new file from them, and the original stays as it was. It runs in your browser, so the document isn't uploaded.
What extracting actually does to the file
A PDF isn't one long stream of text. It's a collection of objects: each page is an object that points to its own content (the drawing instructions for text, lines and images) and to the resources that content uses, like fonts and pictures.
Extracting creates an empty PDF and copies across the page objects you chose, along with everything those pages point to. Things that only other pages use are left behind. That's why an extract of five pages from a 200-page scan is usually a fraction of the size. It isn't always proportional, though: if every page shares one big embedded font or a background image, each extract still has to carry it.
Nothing is re-rendered or converted. Text stays text, images keep their original quality, and links and comments on those pages come along with them.
There are a few things that belong to the document rather than to any page, and those behave differently:
- Bookmarks. The outline is kept for the pages you extracted, unless you tick "Discard bookmarks". A bookmark pointing to a page you left out is dropped.
- Document properties. The new file is a new document, so the original's Title and Author fields aren't copied. If that matters, set them again with Edit metadata.
- Fillable forms. The fields are tied to the document's form structure, which isn't rebuilt. If the pages have filled-in fields, flatten the form first so the answers become part of the page.
Selecting the pages you want
Once the file loads you see a thumbnail of every page. Click a page to select it and click again to unselect. Shift-click selects everything between the last page you clicked and this one, which is the quickest way to grab a run of pages by eye.
For long documents, typing is faster. The box at the top takes ranges like these:
- 88-93 for a single run
- 1, 5, 12-14 for a mix
- 120- for page 120 to the end
- last for the final page
Type the range and click "Select" (or press Enter). The buttons next to it do the obvious things: "All", "Odd", "Even", "Clear", and "Invert", which flips the selection. Invert is handy when it's easier to say what you don't want: select the three pages to leave out, invert, and everything else is selected.
Under the file name you get a running summary such as "Selected 6 of 140: 88-93." Read it before you extract. It catches the classic typo of 8-93 instead of 88-93 better than squinting at thumbnails does.
One thing to know: pages always come out in their original order. If you select 12, then 3, then 7, the new file runs 3, 7, 12. To change the order, extract first and then use Organize pages, or use Merge PDF with a page range, which keeps whatever order you type.
One file or several
The "Output" menu decides how your selection is packaged:
- "One PDF with all selected pages" (the default) puts everything into a single file, named yourfile-extracted.pdf.
- "A separate PDF for each page" gives one file per page, named like yourfile-page-7.pdf.
- "A separate PDF for each range" groups consecutive pages. Select 1-3, 7 and 10-12 and you get three files: pages 1-3, page 7 and pages 10-12.
The per-range option is the one people miss. It's ideal for pulling several exhibits or invoices out of one bundle in a single pass, as long as each one is a continuous run of pages. When there's more than one file, they're downloaded together as a ZIP.
Extracting pages step by step
- Open Extract pages and add the PDF. If it's password protected, you'll be asked for the password first.
- Select pages by clicking thumbnails, or type a range and click "Select".
- Pick an option in the "Output" menu.
- Open "More options" if you want to discard bookmarks or name the files.
- Click the button in the header, which shows the count, such as "Extract 6 pages".
- Preview the result and download.
Changed your mind halfway through selecting? Undo and redo work on the selection, so a stray click doesn't mean starting over.
Naming the output files
"Output file names" in More options takes a pattern built from placeholders. [BASENAME] is the original file name, [FILENUMBER] counts 1, 2, 3, [CURRENTPAGE] is the first page of each file, and [TIMESTAMP] adds the date and time. Adding # signs pads a number, so [FILENUMBER###] gives 001, 002 and so on, which keeps files sorted properly in a folder. A line below the box shows what the first file will be called, so you can check before extracting.
For a single extract I wouldn't bother. For 30 single-page receipts going into an expenses folder, something like receipts-[CURRENTPAGE###] saves a lot of renaming.
When to use a different tool
Extract pages is for keeping a selection. If you think of it the other way round, there are better fits:
- Removing a few pages and keeping the rest: Delete pages. Same result, but you click the pages you don't want.
- Cutting the whole document into pieces, every 10 pages or at chosen points: Split PDF. The guide to splitting a PDF compares the different split modes.
- Breaking a book or manual into chapters: Split by bookmarks, if the file has a proper outline.
Check the page count
After downloading, open the new file and look at the page count before you attach it to anything. If you expected 6 pages and got 5, the selection summary was wrong and you'd rather find out now than from the person who received it. It's also worth a glance at the last page: statements and contracts often carry a "continued on next page" line that tells you the range should have been one page longer.
Also useful: Split PDF · Organize pages
