You can handle this directly in your browser with our free compress pdf tool — nothing gets uploaded.
If you have ever tried to email a 100 MB PDF, copy a 6 GB file onto a FAT32 USB stick, or upload a thick report to a website that only accepts 10 MB uploads, you know the pain. The file is too big and the transfer just stops. The common workaround is to split the PDF into smaller parts, but you usually need each part to stay under a specific size limit.
In this guide we show a practical way to split a huge PDF by file size: work backwards from your target maximum, estimate how many pages fit in one part, use any PDF splitter to cut the file into those page ranges, name the pieces in order, and finally recombine them when you need one file again. The example below uses a 120 MB, 30-page PDF and a 25 MB limit, which matches the classic email ceiling.
Three limits show up constantly:
The fix is the same: break the PDF into pieces that each fall under the limit. The trick is choosing the right number of pages per piece without checking every file by hand.
You usually know two numbers: the total file size and the maximum size per part. From those you get a rough size per page.
size per page = total size / total pages
pages per part = max size / size per page
number of parts = total pages / pages per part, rounded up
Worked example: suppose your PDF is 120 MB and 30 pages long, and your target is 25 MB per part.
So you cut the file into 5 segments of 6 pages each: pages 1-6, 7-12, 13-18, 19-24, 25-30.
Two warnings. First, PDF page weight is not uniform; a vector-dense page can be several times heavier than a plain text page, so the last part of a segment may overshoot the limit. Leave a 10-15% safety margin: if you target 25 MB, aim each part at 22-23 MB, which means using 5 pages instead of 6 in the example above. Second, check whether the PDF has embedded fonts or high-DPI images; recompressing before splitting often shrinks the total enough that fewer parts are needed at all.
Most PDF tools, including Adobe Acrobat, free options like PDFsam, command-line pdftk on Linux or Mac, and online splitters, let you extract or split by page range. You do not need a built-in split-by-size feature; you just feed in the page ranges you computed.
For the 120 MB / 30 page / 25 MB case above, the commands look like this:
for i in 1 2 3 4 5; do
start=$(( (i-1) * 6 + 1 ))
end=$(( i * 6 ))
pdftk input.pdf cat $start-$end output part-$i.pdf
done
Or in a GUI tool, use Split and enter 5 segments of 6 pages each. If your PDF has uneven page weights, split with the 5-page version, giving 6 parts, and accept the extra smaller final piece.
After splitting, open each part and check its size. If any part exceeds the limit, move one page from that part to the next and re-export; one or two passes will converge.
Numbering is not a cosmetic detail; it is what lets you or your recipient reassemble the file later. Use a zero-padded, predictable scheme:
Rules that help:
Recombining is the inverse of splitting: concatenate the parts in order.
pdftk part-01.pdf part-02.pdf part-03.pdf part-04.pdf part-05.pdf cat output final-recombined.pdf
In Adobe Acrobat, use Combine Files and drag the parts in order. Most GUI splitters also have a Merge or Join button; just make sure the order matches part-01 through part-05, not alphabetical sort of a folder that also contains other files.
After merging, verify that the page count equals the original total, the file size is close to the original within re-encoding difference, and the first and last page look correct. If the recipient is not technical, do this merge for them once and hand over the single file.
Why can't I just compress the PDF instead of splitting it?
You can, and you should try it first; lower image DPI, subset fonts, remove metadata. But a 6 GB file made of scanned images often still exceeds a 4 GB FAT32 cap after compression. Splitting is the reliable second move, and it costs nothing.
Does splitting damage the PDF quality?
No. Splitting only reorders existing bytes and cuts the page tree; no re-encoding of the images happens. Recombining in order restores a byte-identical file. The only place quality is affected is if you re-save from a GUI that recompresses; use save as carefully.
Can I split by size directly, without counting pages?
A few advanced tools can estimate and cut, but the page-range method above is portable across any splitter and any OS. If your tool has split by size, use it, then verify each part's size and merge back if a part is too small.
What if the PDF is encrypted or password-protected?
You must unlock it first, otherwise the splitter will refuse to read the pages. Decrypt with the correct password, work on the decrypted copy, and re-apply the password to the final recombined file if needed. Never split a password-protected file you do not own; that is against the law in most jurisdictions.
How many parts is too many to send by email?
Email providers usually cap total attachments per message around 25 MB, so the answer is dictated by your per-part size, not the count. But in practice, more than 6-8 parts becomes inconvenient to track; if you need more, ship a zip via a file-transfer link and send the link by email instead.
That is the whole workflow: estimate pages-per-part from your size limit, cut the file with any splitter using those page ranges, name the pieces in order, and merge them back when one copy is needed. For the 120 MB / 30-page / 25 MB example, the math gives you 6 pages per part and 5 parts; a clean, predictable split that any recipient can reassemble in seconds.
Related reading: batch compress folder of pdfs under size limit, batch rename pdfs by size limit, compress large pdf on iphone without app.