Reorder pages in a PDF
Reordering the pages of a PDF rebuilds the document with the same pages arranged in a sequence you type, such as 3,1,2,4. Each page is copied unchanged into its new position, so content and quality are preserved. The whole rearrangement runs locally in your browser with no upload. Typical uses: fixing pages scanned in the wrong order, moving a cover to the front, or reversing a document scanned back-to-front.
How it works: the page tree defines order, not the file layout
Reading order in a PDF has nothing to do with where page data physically sits in the file. Order is defined solely by the page tree: the document catalog points at a root Pages node whose /Kids array lists pages (or sub-trees) in sequence, and viewers display them in exactly that array order. Reordering a PDF therefore means producing a document whose Kids array lists the same Page objects in a new sequence.
This tool does it the robust way: it parses your comma-separated sequence, validates every number against the actual page count, then copies the pages one by one — in your requested order — into a freshly created document. Copying rather than shuffling in place sidesteps a family of subtle bugs with inherited attributes: in the original file, a page may inherit its size, rotation or resources from a parent Pages node, and naively moving it under a different parent could silently change how it displays. The copy operation resolves all inherited values onto each page first, so every page arrives in the new file carrying its complete, explicit description.
Because the same Page objects are referenced, a sequence that mentions a page twice simply imports it once and points at it from two positions — near-free duplication — while an omitted page is never imported at all. Nothing is rasterized and no content stream is touched, so the operation is lossless: the output differs from the input only in page order and file-structure bookkeeping, and file size stays essentially the same for a straight permutation.
Worked example: un-shuffling a collated booklet scan
An 8-page booklet was pulled apart and scanned as flat sheets, producing a 4.2 MB PDF in booklet-collation order: 8, 1, 2, 7, 6, 3, 4, 5. To restore reading order you enter 2,3,6,7,8,5,4,1 — that is, for each output position, the source page currently holding that content (output page 1 comes from scan page 2, which holds booklet page 1, and so on). The tool rebuilds the file in about a second; the result is 4.2 MB, unchanged to within a kilobyte, with all eight pages now reading 1 through 8. A quick check that the output is right: the page whose printed folio says 1 is now first, and the printed 8 is last.
Reordering PDF pages — frequently asked questions
How do I fix a PDF that was scanned in reverse order?
Enter the sequence counting down, for example 10,9,8,7,6,5,4,3,2,1 for a ten-page file. Feeding a stack face-down is the usual cause, and reversing the order restores the correct reading sequence.
Can I duplicate a page while reordering a PDF?
Yes. Mentioning the same page number twice, like 1,2,2,3, places two copies of that page in the output. Both copies reference the same fonts and images, so the file grows very little.
What happens if I leave a page number out of the sequence?
A page you do not mention is simply not included in the output, which makes reordering double as a way to drop pages. If that is your main goal, a dedicated page-removal tool is more direct.
Does rearranging pages change their content or quality?
No. Pages are copied as complete objects into their new positions without any re-rendering, so text stays selectable and images keep their original compression. Only the order in the page tree changes.
Will printed page numbers on the pages update after reordering?
No. Numbers printed within the page content are part of the artwork and move with their page, so a page labeled 7 stays labeled 7 wherever it lands. Only the physical position in the file changes.