Remove, Reorder or Rotate PDF Pages (Android)

Removing, reordering and rotating pages are structural edits: they change the document's page tree, not the content of the pages, so they cost no image quality and no text sharpness at all. And if you have rotated a PDF before and found the rotation "didn't save", the reason is almost always that your viewer rotated the view rather than writing a new rotation value into the file — those are two different operations that look identical on screen.

On Android, Convertlix's PDF Page Manager reorders, rotates or removes pages and writes the corrected PDF on the handset, so the document never leaves the device — no upload, no server-side copy, nothing to retain. The scope, up front: Android 8.0 or later, sign-in required (there is no guest mode), 25 MB per file on the free tier against a server-set monthly allowance of conversions shown in the app, and the new file lands in Downloads. The rest of this page is the mechanism, which is worth understanding once.

Why your rotation keeps resetting

Every page in a PDF carries a rotation attribute — /Rotate in the file format — whose value is 0, 90, 180 or 270. It tells any viewer how far to turn the page before displaying it. The page content itself is not rotated; the instruction is.

Viewers offer two completely different things under the same word:

• Rotate the view. The viewer turns the page on your screen and remembers nothing. Close the file, or send it to someone else, and it is sideways again. Chrome's PDF viewer, most Android readers and the Preview toolbar all do this by default, and none of them are misbehaving — a reader is not obliged to modify your file.
• Rotate and rewrite the file. The tool writes a new /Rotate value into the page object and saves a new PDF. Now it travels: every viewer, every printer, every portal sees it upright.

This is why the fix feels flaky. If the person you sent it to still sees a sideways page, you rotated the view. If you opened it on a different device and it was fine, then wrong again elsewhere, same answer.

The useful consequence: because a 90° turn is just an attribute change, it costs nothing. No pixels are resampled, no text is re-rendered, and the file size barely moves. That is not true of rotating an image by an arbitrary angle, which does resample every pixel — Straighten a Crooked Photo of a Document covers that distinction, and it matters if the page is tilted 4° rather than turned 90°.

Removing a page: what it does and does not do

Deleting a page removes it from the page tree, so no viewer will render it and no printer will print it. For every ordinary purpose — dropping a blank feeder page, cutting an internal note out of a document you're sending a client, removing the duplicate that the scanner produced — that is exactly what you want.

Two honest caveats:

It usually saves less space than you expect. The page is gone from the tree, but whether the bytes go with it depends on what that page held. A page of text in a born-digital document was maybe 20 KB, so removing it changes nothing measurable. A scanned page holding a full-page image was 800 KB, and removing it is a real saving. If the goal is a size cap, page removal is a rounding error compared with re-encoding the images — see Why Is My Scanned PDF So Large?.

It is not redaction. For a document you are removing pages from and re-saving, the practical outcome is fine. But do not treat page removal as a security guarantee for genuinely sensitive material in a PDF you did not create: PDFs can retain incremental-update history, and the safe method for something that must be unrecoverable is to remove it in the source document and export a fresh PDF. The same goes for removing a part of a page — covering text with a black rectangle leaves the text underneath it, which is how redaction failures make the news.

The reversed-scan problem

If a whole scanned bundle came out in the wrong order, there is usually a mechanical reason worth recognising, because it tells you the fix.

• Everything is backwards. A sheet-feed scanner that outputs last-page-first, or a stack you loaded upside down. The fix is a straight reversal, not a page-by-page drag.
• Pages alternate correct and blank. Duplex scanning of single-sided originals. Remove the even pages.
• All the fronts, then all the backs. You scanned one side of the stack, flipped it, and scanned the other side. The backs are also in reverse order, so this needs interleaving: 1, n, 2, n−1 and so on. It is fiddly and it is the one case where doing it at the scanner again is often faster than fixing it afterwards.
• One page sideways in an otherwise fine bundle. A landscape table someone fed in rotated. That is a single-page rotation, which is the cheap fix described above.

If the pages are spread across several files rather than one, the order problem is a merge problem instead — Merge PDFs Without Uploading Them Anywhere covers assembling them in the right sequence — and if you are still at the capture stage, Scan a Document to PDF Offline (Android) covers getting the order right the first time.

How to fix the pages in a PDF

1. Open the file and look at the whole page list first. Decide every change before you make one: which page goes, which pages swap, which pages turn. Doing it in three passes is how you end up with page 7 in two places.
2. Rotate first, while the numbering still matches your notes. Rotation doesn't change positions, so doing it before any removal or reordering means the page numbers you wrote down still point at the right pages.
3. Remove the pages you don't want, working from the back forwards. Delete page 9 then page 4, not page 4 then page 9 — removing an earlier page renumbers everything after it, and this is where off-by-one errors come from.
4. Reorder last. With the unwanted pages gone, the sequence you're arranging is the final one.
5. Save a new file and check the seams. In Convertlix, open PDF Page Manager, pick the document through Android's file picker, apply the changes and save. Then read the pages either side of every edit, plus the first and last page of the document.
6. Keep the original until the file is accepted. If a reviewer asks for the page you removed, you want it back without a rescan.

When this doesn't work

• Bookmarks and internal links pointing at a removed page. They don't error; they silently stop navigating anywhere, which is harder to notice than a broken link. Check any table of contents after editing a long document.
• Page numbers printed on the page. Reordering changes the page positions, not a "Page 4 of 12" that a designer typed into the footer. After reordering, the printed numbers are wrong and no PDF tool can fix that — it has to be corrected in the source document.
• Interactive form fields. Fields, calculations and scripts are often defined document-wide; removing or reordering pages can leave calculations referring to fields that have moved or gone.
• Password-protected PDFs. Convertlix has no unlock tool, so an encrypted file can't be opened and edited. Ask the sender for an unprotected copy.
• Editing the text or images on a page. None of this touches page content. That is a different class of tool, and for anything substantive the right answer is usually to fix the source document and re-export.

FAQ

Why does my PDF rotation keep resetting?

Because the viewer rotated the display rather than the file. Most PDF readers offer a view rotation that is deliberately temporary — it never modifies your document, so the moment you close the file or send it to someone else the page is sideways again. To make a rotation travel, you need a tool that writes a new rotation value into the page and saves a new PDF.

Does rotating a PDF page reduce its quality?

No, for a 90°, 180° or 270° turn. The rotation is stored as an attribute on the page and the content is untouched, so nothing is resampled or re-encoded and the file size barely changes. Rotating a raster image by an arbitrary angle like 4° is a different operation entirely, and that one does resample every pixel and softens text slightly.

Does deleting a page make the PDF smaller?

Only by as much as that page weighed. Removing a text page from a born-digital document typically saves a few kilobytes and changes nothing you would notice; removing a scanned page that held a full-page photograph can save several hundred kilobytes. If you are trying to hit an upload cap, re-encoding the images inside the file is the lever that moves the number, not page removal.

Does Convertlix upload my PDF?

No. The file is opened through Android's own file picker, the page tree is rewritten on the phone, and the new file is saved to the device — nothing is sent anywhere, and the app has no storage permission to read files you did not hand it. Its network calls carry identity, licence, quota, configuration and payment data, never a document. Sign in once, switch on airplane mode, and it still works.

Is there a free way to reorder or rotate PDF pages without an app?

Yes. qpdf --rotate=+90:3 in.pdf out.pdf rotates page 3 and rewrites the file, and qpdf --pages reorders by listing page numbers in any sequence — free, open source and exact. Ghostscript can do the same less directly. On a Mac, Preview's thumbnail sidebar lets you drag pages into a new order and rotate them with a keystroke, and it saves into the file. LibreOffice Draw opens a PDF and lets you manipulate pages visually. All of these run locally; the reason to reach for a phone tool is that the document is on the phone.

---

If a page is sideways in a file you're about to send, rewrite the rotation into the document rather than turning it on your own screen.