PDF Split: What Actually Happens to Your File When You Split It
Splitting a PDF looks simple β pick pages, get files. But the output can lose bookmarks, drop form fields, and end up larger than the original. Here is what the operation really does, and what to check before you send the result to someone.
A PDF is not a stack of pages. It is a graph of numbered objects β fonts, images, colour profiles, annotations β with a page tree pointing into that graph. Splitting means building a brand new document and copying the objects each selected page depends on. Nothing is re-encoded, so image and text quality is untouched. But anything stored at the document level rather than the page level does not come along for the ride, and that is where the surprises live.
This tool runs entirely in your browser using pdf-lib. Your file is never uploaded, which also means the work is done by your own CPU β a 500-page scan will take a few seconds and a lot of memory.
Quality is preserved, but the file is rebuilt
A common claim is that split output is byte-identical to the source. It is not, and the distinction matters. Page content streams are copied verbatim β no recompression, no resampling, no generation loss, so a 300 DPI scan stays a 300 DPI scan. But the output file is written from scratch: object numbers are reassigned, the cross-reference table is rebuilt, and unused objects are dropped.
The practical consequence is that a checksum of the split page will not match anything in the original. If you need to prove a document was not altered, split files cannot carry that proof. Keep the original and use a detached signature instead.
What does not survive a split
These live at the document level, so they are not attached to any single page and cannot be copied with one:
- β’Bookmarks and the outline tree β the sidebar navigation disappears entirely
- β’Form fields (AcroForm) β filled values and the interactive fields themselves are dropped
- β’Internal links that pointed to a page outside the range you extracted β they become dead
- β’File attachments embedded in the document
- β’Document-level JavaScript and named destinations
- β’Digital signatures β always invalidated, since the byte range they signed no longer exists
Why the pieces can add up to more than the whole
This catches people out constantly. Split a 10 MB, 100-page report into 100 single-page PDFs and the total can easily exceed 40 MB. The reason is shared resources: in the original, one embedded font and one company logo are stored once and referenced by all 100 pages. After splitting, every output file needs its own copy of both.
The effect scales with how much each page shares. A text-heavy document with embedded fonts inflates badly. A folder of scanned photographs, where each page already carries its own unique image, barely changes at all. If your split output is unexpectedly large, this β not a settings mistake β is almost always why.
Split, extract, or remove β three different jobs
These get confused often, and picking the wrong one means redoing the work:
- Split
- One input becomes many outputs. Use it when you want every page, or every chunk, as its own file β for example turning a batch scan back into individual invoices.
- Extract
- One input becomes one smaller output containing only the pages you named. Use it when you need to send chapter 4 and nothing else. Set this tool to a page range to get exactly this.
- Remove pages
- One input becomes one output with certain pages deleted β the inverse of extract. Reach for it when the pages you want to drop are fewer than the pages you want to keep, such as stripping a cover sheet.
Getting the page ranges right
- 1
Page numbers are the ones you see
Ranges are 1-based and inclusive, matching the page numbers your PDF reader shows in its toolbar. Entering 5-10 gives you six pages, not five. Note that this is the physical page position in the file, which may differ from a printed page number if the document has unnumbered front matter.
- 2
Combine ranges and single pages
You can mix them in one expression: 1, 4-7, 12 pulls out six pages into one file. Order does not matter for selection, but pages always come out in their original document order.
- 3
Decide between one file and many
Extracting a range produces a single PDF and downloads directly. Splitting into individual pages produces many files, so they are bundled into a ZIP archive automatically. There is no way to download twenty separate PDFs at once without the archive β browsers block that.
- 4
Check the output before sending
Open the result and confirm the pages are the ones you meant, especially if the document has front matter that shifts the numbering. If the original had bookmarks or form fields you needed, this is where you will notice they are gone.
Extended FAQ
Will splitting reduce image quality?
No. Pages are copied as-is with no recompression, so a scan stays at its original resolution and JPEG artifacts are not compounded. You can split the same file repeatedly without any generation loss.
Can I split a password-protected PDF?
It depends on which password. PDFs with an owner password β the kind that only restricts printing or copying β will open and split normally. PDFs with a user password, which is required just to view the file, cannot be read without that password. Either way, the output files are not encrypted: any restriction on the original is gone, so do not assume the pieces inherit the parent's protection.
Why did my bookmarks disappear?
Bookmarks are stored in a document-level outline tree, not on individual pages, so there is nothing page-specific to copy. Every browser-based splitter behaves this way. If the outline matters, split with a desktop application that explicitly supports rebuilding it, or add bookmarks again afterwards.
Is there a page or file size limit?
There is no fixed limit, but the whole document is held in your browser's memory while it works, so very large files are bounded by your device rather than by us. Several hundred megabytes is realistic on a desktop; a phone will struggle much sooner. If the tab becomes unresponsive on a huge file, close other tabs and try again.
Are my PDFs uploaded to a server?
No. The file is read and processed by JavaScript in your browser, and the output is generated locally. Nothing leaves your device, which is why it also works with no network connection once the page has loaded.
How do I rejoin the pieces later?
Use a PDF merge tool and add the files in the order you want. Merging is lossless in the same way splitting is, but it will not restore the bookmarks or form fields that were lost on the way out.