What this does
It removes the pages you name from a PDF and hands back the document without them: the blank scans, the wrong-printer test page, the one page that should not leave the building. You list pages the way a print dialog does, see exactly how many will go and stay, and download a file that is your document minus those pages, name, author and dates intact.
Everything happens in your browser. The file is read off your machine, rebuilt on your own processor, and handed back; nothing is uploaded, which matters twice over for this tool, because the page being deleted is so often precisely the sensitive one.
How to use it
- Drop your PDF onto the tool, or press “Choose a PDF”. The page count appears once the file has been read.
- Type the pages to delete:
7is a page,2-5is an inclusive range,12-runs to the end,-3means from the start, commas between. The line underneath answers back in plain words, “Removes 3 of 12 pages, keeping 9”, or explains exactly what it could not read, including the document’s real page count. Deleting every page is refused there too, before the button rather than after. - Press “Delete pages”, then download. The file comes back as yourname-pages-removed.pdf, original untouched.
Deleted means gone
There are two ways to remove a page from a PDF, and they differ in the way that matters. The quick way edits the document’s page list so the page no longer displays, and leaves the page’s actual content, its text, its images, sitting orphaned inside the file. We measured it: a ten-page file that lost nine of its pages that way shrank from 13,182 bytes to 13,164. Everything “deleted” was still in the file, recoverable by anyone with a reason to look, in a document whose whole point was that those pages should not travel with it.
This tool takes the other way: it rebuilds the document from the pages you keep, so removed content is not hidden from view but absent. The test suite proves the property directly, by planting a known image on a doomed page and checking its bytes exist in the input and do not exist in the output.
What survives, and what it costs
Kept pages come through exactly, as objects: fonts, images, sizes, selectable text, all untouched. The document’s identity, its title, author, subject and creation date, is carried onto the rebuilt file, because the result is still your document, just shorter.
What the rebuild cannot carry is the catalog layer: bookmarks and fillable form fields do not survive, for the same structural reason as in the merger and splitter. Where an edit genuinely does not need a rebuild, this site prefers the in-place kind, and the page rotator is exactly that; deletion is the case where in-place editing quietly betrays the point, so here the trade goes the other way, and the page says so rather than letting you find out.
When it matters
Redaction-by-removal is the big one: the pricing page from the proposal going to a different client, the salary schedule from the contract, the notes pages from the board pack. Scanned bundles collect blank backs of pages and accidental duplicates that want removing before filing. And forms ask for “pages 1 and 3 only” often enough that removing 2 is a job in itself.
One caution belongs here rather than in small print: removing a page removes that page’s content, not mentions of it elsewhere. If page 2’s figures are quoted on page 5, deleting page 2 does not unquote them. Deletion is honest removal, not redaction review.
Limitations
You need to know the page numbers. This tool does not show thumbnails, so check the document in your viewer and bring the numbers; the preview line confirms the arithmetic before anything is deleted.
The interactive layer is not carried. As above: bookmarks and fillable fields do not survive the rebuild that makes deletion real.
Protected files are refused. The same door policy as this site’s other PDF tools: remove the password first in software that has it, then delete the pages.