Drop a PDF into the zone below and every page becomes a tile you can drag, turn or delete. Move page 40 to the front, put the scan that came out upside down the right way up, drop the duplicate cover sheet, then rebuild the document in one pass. The number in each corner is the page number it will have when you are finished, not the one it started with. Nothing is uploaded. The original file is never modified.
Your files are read by JavaScript running on your own machine. There is no upload, no queue, and no copy on our server to delete later.
Drag a card to reorder it, or use the arrows under each card. The order here is the order in the output. Each card is a preview of the finished page, so what you see is what the PDF will contain.
Esc close ← → move R rotate
Starting
This is your processor doing the work, so the time it takes depends on your machine and not on how many other people are on the site.
The board above does the job without any of this. The guide below is for the part that bites afterwards - why a rotated page can print correctly and still look wrong on screen, what happens to links and bookmarks when page 4 becomes page 19, why reordering never changes the file size, and how to fix a document that was scanned in the wrong order without rescanning it.
Two minutes for a normal document, nothing to install, and no account for your first run. This PDF page organizer runs inside the page you are reading, on your own processor.
Click the drop zone above, or drag the file onto this page. Every page is rendered as a tile. Tiles are drawn as you scroll rather than all at once, which is why a 600 page document becomes usable straight away instead of after a minute of waiting.
Grab a tile and drop it where it belongs. On a touch screen use the two arrows underneath each tile instead, because browser drag events are unreliable on phones and a tool that only supports dragging does not work on half its traffic.
Each tile has a rotate left, a rotate right and a delete button. Deleted pages stay on the board, faded and struck through, with an undo button where the delete button was, so a mis-tap costs one click rather than a restart. The bar above works on everything at once. Turn all right, Turn all left, Reverse order, Undelete all and Start over.
The panel on the right reports pages in, how many were deleted, how many were rotated, whether the order changed and how many pages come out. Press Organize PDF and the rebuilt document downloads immediately.
Keyboard shortcuts work on any focused tile: R turns it right, Shift+R turns it left, Delete removes it and Alt with the arrow keys moves it. For a fifty page reordering job that is considerably faster than dragging, and it does not scroll the page out from under you.
Rotation in a PDF is not what most people assume, and the assumption causes a specific, very annoying bug: a page that looks correct in one viewer prints sideways from another.
When a scanner produces a page the wrong way up, the pixels are stored exactly as they came off the sensor. The scanner does not turn them around. It writes a /Rotate entry on the page instead, saying "display this at 90 degrees". The rotation is an instruction to the viewer rather than a change to the content, and every compliant viewer and every compliant printer honors it, which is why this is a legitimate way to store a rotated page.
This tool works the same way, deliberately. Turning a tile adds a quarter turn to whatever /Rotate the page already carried, then writes the result. The image inside the page is never re-rendered, never re-compressed and never touched. A 300 dpi scan turned upright here is still the identical 300 dpi image it was before, with a single number changed on the page dictionary.
The alternative - decoding the image, physically rotating the pixels and re-encoding - would produce a file that looks the same, is slower to make, is larger, and has been through a generation of loss for nothing. It is what several online tools do, and it is why a scan that has been "fixed" twice looks noticeably worse than the original.
The rotation is added to what is already there, not applied instead of it. A page stored at 90 degrees that you turn once here ends up at 180, not at 90. That is why turning a page four times returns it exactly to where it started, and why a document with mixed rotations can be straightened by turning each odd page individually rather than by rotating everything at once.
PDF pages live in a tree. The page objects themselves are scattered anywhere in the file, and the tree is what records the order they are presented in. So reordering rewrites the tree and leaves the pages alone, which is why moving page 40 to the front does not move a single byte of page 40's content.
This tool rebuilds the document rather than editing it in place: a new file is created and the pages are copied into it in your order, with each page bringing its content stream, fonts, images and annotations across as compressed data that is never decoded. What you download is structurally new and visually identical, page for page, to what you put in.
That is why reordering never changes the file size in any meaningful way. Nothing was added and nothing was removed. If you also deleted pages, the file shrinks by whatever those pages exclusively owned - which on a scan is a lot and on a text report is very little, because fonts and letterheads are shared with the pages you kept.
Rearranging pages is a bigger change to a document than it looks, because several things in a PDF refer to pages by position.
Reordering does not renumber anything printed on the pages. If the original had "Page 3 of 12" in the footer, that text is part of the page artwork and moves with it. A document you have reshuffled will have footers that disagree with the new order, and the fix is Add Page Numbers on the finished file, not this tool.
The specific situations people arrive here with, and the fastest route through each:
Deleted tiles stay on the board, faded and struck through, with an undo button where the delete button was. Nothing is committed until you run the operation, and the source file on your disk is never written to at any point.
That design costs a little screen space and buys the thing that matters most on a touch screen: a mis-tap is an inconvenience rather than a restart. The Undelete all button in the bar restores every deleted page at once, and Start over resets order, rotation and deletions back to the document as it arrived.
Reordering is a low-drama operation performed on high-drama documents. Signed contracts, case bundles, board packs. Medical files being assembled in the right sequence before they go to a consultant for a referral. The content is often the most sensitive thing an organization owns, and the change being made to it is completely trivial, which is exactly what makes uploading the whole bundle feel so disproportionate.
Uploading a complete case bundle to a third party server so that page 12 can move to position 3 is a poor trade. This tool has no server in the path. The file is read by JavaScript on your own machine, the new document is assembled in your own memory, and it is handed back to you as a download. There is no queue, no temporary directory, and no deletion policy for you to trust.
Prove it without believing a word of the above: load this page, disconnect from the internet completely, then drop a PDF in and reorganize it. It works. The only request a run makes counts your operation against the free allowance, and it carries the name of the tool and nothing about the document you just rearranged.
Every page thumbnail on the board was drawn on your own processor. A server-side organizer has to render those previews somewhere, which means it has read, rasterized and cached a picture of every page of your document before you have made a single change to it.
The board renders page tiles as they scroll into view rather than all at once, so a 600 page document is usable within a second or two and only the pages you actually look at are ever rasterized. That handles the rendering cost.
What it does not handle is a genuinely enormous file: the source document is held parsed in memory while the new one is assembled alongside it, so peak usage is roughly twice the document. The stated ceilings are 100 MB on the free tiers and 500 MB on Premium. A failed run is refunded automatically. Dragging tiles is also simply the wrong interface past a few hundred pages - if you are reordering a 900 page bundle, split it, organize the parts, and merge them back.
| Symptom | Cause | Fix |
|---|---|---|
| A rotated page still looks wrong in one program | That program is ignoring the page's /Rotate entry, which is a fault in the viewer rather than in the file. | Check the file in a second viewer or print a test page. Compliant software honors it. |
| Turning a page went the wrong way | The page already carried a rotation, and yours was added to it. | Correct, and intended. Keep turning: four presses always return a page to where it started. |
| The footers say Page 3 of 12 in the wrong places | Printed page numbers are artwork and move with their page. | Stamp fresh pagination on the result with Add Page Numbers. |
| Links in the document stopped working | They pointed at pages that moved. | Unavoidable. A broken link is dropped rather than left pointing at whatever is now in that position. |
| Dragging does nothing on a phone | Browser drag events are unreliable on touch screens. | Use the two arrow buttons under each tile. They do the same job and always work. |
| The file is exactly the same size afterwards | Reordering moves pages between positions and adds nothing. | Working as designed. Only deleting pages changes the size, and then only by what those pages exclusively owned. |
Drop the PDF into the zone at the top of this page. Every page appears as a tile. Drag a tile to move it, or use the arrows underneath it on a touch screen. The number in each corner is the position that page will have in the finished document. Press Organize PDF and the rebuilt file downloads. No account is needed for your first run and there is no watermark.
No. The document is read by JavaScript on your own machine, every page thumbnail is drawn on your own processor, and the new file is assembled in your own memory. Load this page, disconnect from the internet, and it still works. A server-side organizer would have to rasterize a picture of every page of your document before you made a single change to it.
Each tile carries a rotate left and a rotate right button, and a focused tile responds to R and Shift+R on the keyboard. The rotation is written as the page's /Rotate value, so the image itself is never re-rendered or recompressed - a 300 dpi scan turned upright here is still the identical 300 dpi image, with one number changed.
Because the page already carried a rotation and yours was added to it. A scanner that produced a sideways page recorded 90 degrees on it, so your quarter turn makes 180. That is correct. It is also why turning a page four times always returns it exactly to where it started.
Press Reverse order in the bar above the board. Every page flips end for end in one click, which is the exact inverse of what a document feeder does when the stack goes in face down. Check the first and last tiles. Then rebuild.
Neither. Pages are copied as compressed streams and are never decoded or re-encoded, so the content is bit-identical and the size barely moves. Only deleting pages changes the size, and then only by whatever those pages exclusively owned - which on a scan is a lot and on a text report is very little, because fonts and letterheads are shared with the pages you kept.
Bookmarks are not carried into the output, because an outline built for the old order would point at the wrong pages, and navigation that silently goes to the wrong place is worse than navigation that is absent. Internal links to moved pages are dropped for the same reason. Everything else survives. Page content, rotation, page sizes and form fields all travel with their pages.
No. No tool that reorders pages can do this, because text reading "Page 3 of 12" in a footer is artwork drawn on that page, indistinguishable from any other text on it, and it moves with the page. Stamp fresh pagination onto the finished document with Add Page Numbers.
Yes. That is what this tool is for: delete, rotate and reorder all happen on the same board and are written to the new document in a single pass. Deleted pages stay visible but faded, with an undo button in place of the delete button, so a mis-tap costs one click rather than a restart.
Not directly. An encrypted PDF cannot have its pages read or rendered, so no thumbnails can be drawn and the file is reported as locked rather than silently failing. Remove the protection with Unlock PDF first, on a document you are authorized to open.
Not in one step. This tool takes one document, which keeps the board unambiguous about where each page came from. Run Merge PDF first to combine the files, then bring the merged document back here and drag the pages into whatever order the job actually needs.
The board handles it: tiles render as they scroll into view rather than all at once, so only the pages you actually look at are ever rasterized. Whether that is wise is another question. Dragging tiles stops being a sensible interface somewhere in the low hundreds, so for a very long bundle, split it, organize the parts, and merge them back.
No. What you download holds your pages in your order and nothing else - no stamp, no logo in the corner, no trial banner, and no attribution line you are obliged to keep. It is free for commercial use.
Never. Your source file is opened read-only and nothing is ever written back to it. What you download is a separate document, built from the pages you kept in the order you set. Get the order wrong and the original is untouched, so you can drop it in again and start from a clean board.
All of these run the same way: your file is read by JavaScript on your own machine, the work happens on your processor, and the result is handed back to you as a download that never existed anywhere else.
Find the apps that slow your store down, by name and in milliseconds, then answer shoppers' doubts with trust badges under Add to cart. Both are free, with no card and no access to your store admin.
Paste any page and get its three Core Web Vitals, the apps costing time, like Klaviyo or Judge.me, and the Shopify or WooCommerce fix.
Show the promises that settle doubts, from free shipping to a money-back guarantee, right under the Add to cart button. One line in your theme.
Change the link after printing and count the scans. 5 tracked codes on a free account, static codes unlimited.
Anonymous visitors get one operation per 12 hours, per network address. That is not a sales tactic, it is the only lever we have: this tool runs on your processor, not ours, so the run itself earns us nothing. An account tells us who you are, and the ceiling goes up the moment you have one. Your allowance resets in 12 h.
No credit card, no trial period, no sales call. Join with Google, GitHub, Facebook or email.
Free accounts get two operations per 12 hours, and yours reset in 12 h. Waiting costs you nothing and is a completely reasonable choice. If PDF work is a weekly job rather than a one-off, the counter is the wrong tool for you and Premium takes it off the page.
Cancel anytime. Everything you have already produced stays yours either way.