How to split large PDF files
By page, by groups or by ranges, with names that keep the order
The three ways to split a PDF, how to write page ranges that do what you mean, how the parts are named so they stay in order, and how to split a document that is too big for an upload or attachment limit.
- By HelpMeJPG editorial team
- Published
- Updated
- 9 min read
On this page
Quick answer: Open the document in the Split PDF tool, choose whether you want one file per page, fixed groups of pages or your own ranges such as 1-3, 4, 5-8, check the thumbnails, and press Split. Each part is a complete PDF containing only its pages, copied without re-rendering, and the parts are named with a part number and page range so they sort back into the original order. If you are splitting to get under an upload limit, compress the document first and then divide it into as few parts as the limit allows.
Three ways to split a PDF
Splitting always does the same thing underneath – it creates new, empty documents and copies the chosen pages into them – but there are three ways to say which pages go where, and picking the right one saves a lot of typing.
| Mode | What it produces | Best for |
|---|---|---|
| Every page | One file per page, so a 40-page document gives 40 files | Batch scans of separate receipts or forms; feeding pages into another system one at a time |
| Every N pages | Files of N consecutive pages each; the last file takes whatever is left | Getting under a size limit; splitting a book of identical two-page or four-page records |
| Custom ranges | One file per range you write, in the order you write them | Chapters, appendices, exhibits – anything where the boundaries are known |
"Every page" and "every N pages" need no further instructions. Custom ranges are where mistakes happen, so the next section is about writing them correctly.
Try it: Split PDF – choose one of the three modes and check the summary line before splitting.
The thumbnails are not decoration. Before pressing Split, scroll to each boundary and confirm that the last page of one part and the first page of the next are where you expect them – a cover sheet, a blank separator page or a page numbered "1 of 3" is usually the clue. The summary line tells you how many files you are about to create; if it says 40 when you expected 4, you have probably used "every page" where you meant "every 10 pages".
Writing ranges that do what you mean
Ranges use the same notation as a print dialog. Page numbers count from 1 and refer to the physical position in the file, not to any number printed on the page. Separate ranges with commas; each comma-separated item becomes one output file.
| You type | You get |
|---|---|
1-3 |
One file with pages 1, 2 and 3 |
1-3, 4, 5-8 |
Three files: pages 1–3, page 4 on its own, pages 5–8 |
5- |
One file from page 5 to the end of the document |
-3 |
One file with pages 1 to 3 |
1-3, 7-9 |
Two files; pages 4–6 are left out entirely |
That last row is important: pages you do not mention are not included anywhere. This is a feature when you want to pull out two chapters and ignore the rest, and a trap when you have simply forgotten a range. Before splitting, add up the pages your ranges cover and compare the total with the document's page count.
Ranges are validated strictly. A page listed twice, a range that runs backwards (8-5), or a page beyond the end of the document is reported as an error rather than silently fixed, because every page can belong to exactly one output and a guessed correction is worse than a clear message. If you need the same page in two parts – a cover sheet in front of every chapter, say – split first, then merge the cover into each part.
Naming and keeping the parts in order
Each output file is named after the original, followed by a part number and the pages it contains: a file called annual-report.pdf split into three ranges produces annual-report-1-pages-1-3.pdf, annual-report-2-page-4.pdf and annual-report-3-pages-5-8.pdf. When there are ten or more parts the part number is zero-padded (01, 02 … 12), so the files sort correctly in any file manager, email client or upload dialog. Names come from the file name you supplied, never from the document's metadata title, so you always know where a part came from.
A few habits make a split set easier to live with:
- Rename the source before splitting. The output names inherit it, so
smith-contract-signed.pdfgives parts that explain themselves, whereasscan0001.pdfdoes not. - Download the ZIP when there are more than a handful of parts. One archive is easier to move around than thirty files, and the files inside keep their names and order.
- Keep the original. Splitting never modifies it; keep it as the reference copy and treat the parts as derived files.
- Do not rename parts by hand to match printed page numbers. The part names describe physical positions in the file, which is what a merge needs later. If the printed numbering matters, put it in a covering note instead.
Reassembling is straightforward: add the parts to the merge tool, where they will already be in the right order if their names sort correctly, and merge. The guide on how to merge PDF documents covers the details, including what happens to bookmarks.
Splitting to get under upload and attachment limits
The most common reason to split a PDF is that something refuses to accept it whole. Email is the classic case – Google's documentation puts the limit for personal Gmail accounts at 25 MB per message – and portals for job applications, tax returns, tenders and court filings often impose their own, sometimes much lower, ceilings.
Two facts shape the strategy. First, splitting does not reduce the total: ten parts of a 60 MB scan still add up to 60 MB, so if the problem is total storage rather than a per-file cap, splitting does not help. Second, bytes are not spread evenly across pages. In a scanned document every page costs about the same, so you can divide the file size by the page count, work out how many pages fit under the limit, and use "every N pages". In a born-digital report, text pages are almost free and a few pages with large photographs or maps carry most of the weight; a split by page count can then produce one part that is still too large. Check the size of each part after splitting, and move a boundary if one part is out of proportion.
Email adds a complication of its own: attachments are encoded in a way that makes them roughly a third larger in transit, so a part that is just under a stated limit on disk can be rejected. Leave headroom – aim well below the limit rather than right at it.
Usually the better first move is to make the whole document smaller before dividing it. Run it through the Compress PDF tool and look at the result; a scan often shrinks enough that no split is needed, and if a split is still needed it will take fewer parts. Compressing after splitting works too, but means judging the quality of several files instead of one. Preparing PDFs for email walks through the combined decision, and how to reduce PDF file size explains what the compression profiles do to the pages.
Splitting, extracting and deleting: which tool when
Three tools on this site take pages out of a document, and they overlap. The distinction is what you want to end up with.
| You want | Use |
|---|---|
| Several new files that together cover the document | Split PDF |
| One new file containing a selection of pages | Extract PDF Pages |
| The original document minus a few pages | Delete PDF Pages |
| The same pages in a different order | Reorder PDF Pages |
Extraction keeps pages in the order you select them, which makes it a quick way to build a short excerpt; splitting keeps every part in document order. Deletion refuses to remove every page, so you cannot produce an empty file by accident. The guide on arranging and reordering PDF pages covers the extraction and deletion tools in more detail.
If the reason for splitting is confidentiality – sending someone only the pages they are entitled to see – splitting or extracting is the right approach, because pages that are not copied are genuinely absent from the new file. Cropping a page or covering content is not: it hides material that remains in the file. Handling confidential PDFs safely goes through the difference.
What survives a split and what does not
Because pages are copied rather than re-rendered, each part is a faithful subset of the original. Page size, orientation and rotation are preserved; text stays selectable; embedded images and fonts are identical bytes; links, comments and form widgets that sit on a page come with it. Each part also receives the original document's title so viewers show something sensible in the title bar.
Document-level structures do not come along. Bookmarks – the outline shown in a viewer's side panel – point at pages by position in the original and are not rebuilt in the parts. Custom page labels (front matter numbered i, ii, iii) are dropped, so viewers number each part from 1. File attachments and document-level scripts are not carried over. If the printed page numbers in a part should match the original, they already do, because they are part of the page content; if a part needs its own numbering, Add Page Numbers can stamp it on afterwards.
Password-protected documents must be opened with their password before they can be split, and the parts are written without protection. If they should stay protected, add a password to each part afterwards with Protect PDF.
Very long documents
The tool handles documents up to 2,000 pages and produces at most 500 files from one split. A 2,000-page archive therefore cannot be split into single pages in one pass – nor should it be; files of 5 or 10 pages are more manageable. If you genuinely need one file per page from a very long document, split it into blocks of a few hundred pages first, then split each block. Scanned documents take longer than text documents to split because every page carries a large image, but the work still happens entirely in your browser and the page remains responsive while it runs.
Frequently asked questions
Does splitting a PDF reduce its quality?
No. Pages are copied into the new files exactly as they are, without being rendered to images, so text, fonts, vector graphics and embedded pictures are byte-for-byte identical to the original. The only things that change are document-level extras such as bookmarks, which are not rebuilt in the parts.
How do I split a PDF so each part is under an email limit?
Compress the document first, then divide its size by its page count to estimate how many pages fit comfortably under the limit, and split with the "every N pages" mode using that number. Check the size of every part afterwards, because pages with large images weigh far more than text pages, and leave headroom since email encoding makes attachments larger in transit.
What happens to pages I leave out of my custom ranges?
They are not included in any output. Custom ranges produce one file per range you type and nothing else, which is convenient for pulling out two chapters but means a forgotten range silently loses pages. Add up the pages your ranges cover and compare the total with the document's page count before splitting.
Sources and further reading
About the author
HelpMeJPG editorial teamEditorial team
The people who build and maintain the HelpMeJPG tools. Guides are written from the tools' own behaviour and the standards they implement, and are revised whenever a tool changes.