A PDF that will not send is one of the most common document problems there is, and there are two fixes: make it smaller, or cut it up. This guide covers when each is right, the actual limits you are fighting, and how to split a PDF into parts that each stay under a cap without guessing.
How to split a PDF by file size
- Open ihatepdf.cv/split-pdf-by-size — no account, nothing to install.
- Drop in the oversized PDF. It is read in your browser and never uploaded, so there is no size cap.
- Choose a limit — use a preset for Gmail or Outlook, or type your own in megabytes.
- Split. Each part is built and then measured, and rebuilt smaller if it came out over.
- Download the parts individually or all at once as a ZIP.
The limits you are actually fighting
Gmail refuses attachments over 25 MB and substitutes a Drive link. Outlook.com stops at 20 MB. Corporate Exchange is frequently configured far lower — 10 MB is common and 5 MB is not rare, because the limit is set by an IT policy rather than by the product default. Government and university submission portals are usually the strictest of all, often capping uploads at 2 to 5 MB.
There is a catch that catches almost everyone. Email attachments are base64-encoded for transport, which inflates them by roughly a third. A 24 MB file can therefore still bounce off a 25 MB limit, because what travels is closer to 32 MB. Leave ten to twenty per cent of headroom below the stated cap and the problem disappears.
Split or compress? Compress first
If the document is large because of scans or photographs, compression will often bring the whole thing under the limit and keep it as a single file — which is what recipients actually want. Scanned documents commonly lose 50 to 70 per cent with no visible difference, because the bulk is image data with far more resolution than a screen or an office printer can use.
Split when compression cannot get you there, when the document is genuinely enormous, or when the recipient wants it in sections anyway — a hundred-page appendix is often more useful as five files than one. Text-heavy documents are the case where splitting wins early, because text is already vector data and barely compresses at all.
Why splitting by page count does not work
The obvious approach is to divide the file size by the page count and cut every N pages. It fails on almost every real document, because pages differ in weight by two or three orders of magnitude.
A twenty-page report with four full-bleed photographs might carry ninety per cent of its bytes in those four pages. Split it evenly and one part sails over the limit while the others are nearly empty — and you only find out when the email bounces. Splitting properly means building each part and then measuring it, rebuilding smaller if it came out heavy. That is slower by a second or two and correct every time.
What happens to quality
Nothing. Splitting copies pages across exactly as they are — nothing is re-encoded, re-sampled or rasterised, and the parts together contain precisely the original pages. This is the practical difference from compression, which trades image quality for size. If you need the document whole again, Merge PDF puts the parts back together in order with nothing lost.
When one page is bigger than the limit
Occasionally a single page — a full-bleed poster, a high-resolution map, a photographic plate — exceeds the cap on its own. It cannot be divided further, so any honest tool has to hand it back oversized and say so. The fix is to compress that page before splitting, or to accept that one part will need a file transfer link.
Sending the parts so they actually arrive
A few habits save a round of confused replies. Name the parts so they sort correctly — document-01, document-02 — which is what any sensible splitter does automatically. Say in the email how many parts there are; the commonest failure with a split document is not a corrupt file but a recipient who never realised part three existed. Send them in separate emails rather than one, since several attachments in one message hit the same total limit you were trying to avoid. And if the recipient needs it as one document, tell them the parts merge back cleanly.
Why there is no size limit here
Most online splitters cap the file you can upload, because every megabyte costs them bandwidth and processing. That cap is frequently lower than the document you are trying to fix, which is a particularly frustrating way to fail. Splitting in the browser has no such economics: your file never leaves the machine, so the only ceiling is your own device memory, which on a desktop comfortably covers documents of a few hundred megabytes.
Frequently asked questions
Should I split or compress a PDF that is too large for email?
Compress first if it is image-heavy — it usually works and keeps the document as one file. Split when compression cannot reach the limit or when the recipient wants sections.
What size limit should I set?
Below the stated cap, not at it. Twenty megabytes for Gmail and fifteen for Outlook are safe working numbers, because base64 encoding inflates attachments by about a third in transit.
Why do the parts have different page counts?
Because pages are not the same weight. Parts are built to a byte budget, so six text pages and two scanned pages can come out the same size.
Does splitting reduce quality?
No. Pages are copied untouched; nothing is re-compressed or rasterised.
Can I merge the parts back later?
Yes, with Merge PDF, and nothing is lost — splitting never altered the pages.
Is there a maximum file size?
No server limit, because there is no server. The practical ceiling is your device memory.
What if one page is larger than my limit?
It is written as its own part and flagged as over. Compress that page if it has to fit.