PDF Too Large to Email? Work Out Which Fix You Need
The message bounced. The attachment is too large, and you have a number to get under.
This is a different problem from wondering why a file is big. You have a hard limit and a deadline, and the fastest route to meeting it depends entirely on what kind of document you are holding — with the additional irritation that the most obvious remedy, compression, is the right answer for roughly one case in three.
The limit is smaller than the number
Start here, because it explains failures that otherwise make no sense.
Email cannot carry raw binary data. Attachments are re-encoded into text for transit, and that encoding is 33% larger than the file it encodes — measured, not estimated: three bytes become four characters, every time.
So the advertised limit is not the file size you can send:
| stated limit | file size that actually fits |
|---|---|
| 10 MB | about 7.5 MB |
| 25 MB | about 18.8 MB |
| 50 MB | about 37.5 MB |
A 24 MB attachment against a 25 MB limit becomes roughly 32 MB in transit and bounces. Nothing is broken; you were measuring the wrong thing.
Aim for three quarters of the stated limit and you will stop being surprised.
Three limits apply, and the smallest one wins
Your provider has a sending limit. Your recipient’s provider has a receiving limit. If they work somewhere with a mail gateway, that has its own, often stricter.
You can see the first. You cannot see the other two, and a message can leave your outbox successfully and be rejected further along — sometimes silently, which is the worst version.
When it matters, ask. “What is the largest attachment you can receive?” is a faster conversation than three failed sends.
Which route do you need?
The five-second test decides most of this. Try to select the text on a page.
Nothing highlights — the pages are images. It is a scan, and compression will genuinely work. Route A.
Text highlights normally — the file contains real text, and compression is probably the wrong tool. Routes B to E.
That distinction is the whole diagnosis, and it is explained in full in why is my PDF so large.
Route A: it is a scan — compress it
For image-based documents, re-encoding pages at a lower resolution is the only thing that meaningfully reduces size, and the reduction is usually large, because scanners default to far more resolution than a document page needs.
Compress PDF does this in your browser. Pick a quality level, run it, and check a page or two afterwards — particularly small print and figures — before sending.
If you still have the paper and the scan was made at an unnecessary setting, rescanning at 300 DPI greyscale beats compressing a bad scan. That, and the arithmetic behind why 300 DPI greyscale is the right default, is covered in how to scan and file paper records.
Route B: it has real text — go back to the source
Do not compress a text-based PDF. It converts your text into pictures of text: search stops working, the document becomes unreadable to screen readers, and the file typically gets larger, because characters are cheap and photographs of characters are not.
If the document has real text and is still huge, the weight is almost always images placed inside it at full camera resolution. The fix belongs in the source document — resize the images, then export again using whatever “reduce file size” or “screen quality” option your software offers.
Exporting the same document twice, once for print and once for sending, takes seconds and is frequently the difference between 40 MB and 4 MB.
Route C: send only what they need
The most underrated answer, and often the best one.
A twelve-page contract is regularly three pages of substance and nine of schedules. If your recipient asked about the payment terms, send the payment terms. Extract PDF Pages pulls out exactly the pages you name — see how to extract specific pages.
This is better than compression on every axis. It is faster, it degrades nothing, the recipient reads less, and it discloses less. The only reason it is not the default answer is that it requires thinking about what the other person actually needs.
Route D: split it into parts
When they genuinely need all of it and it will not shrink, divide it. Split PDF produces several files, each small enough to send — how to split a PDF.
Two things make this tolerable for the recipient rather than infuriating:
Name the parts so the order is obvious. report-1-of-3.pdf, not report-a.pdf. The person on the other end has to reassemble this mentally.
Split at sensible boundaries — chapters, sections, exhibits — rather than at arbitrary page counts, so each file is a thing rather than a fragment.
Say in the message how many parts there are. A recipient who does not know part three exists will not ask for it.
Route E: send a link instead
Upload it to a storage service and share the link. This is the standard answer in most organisations, and it has real advantages: no size limit worth worrying about, and you can see whether it was opened.
It also has a real cost, and it is the one this site exists to take seriously: the document is now on someone else’s infrastructure. For a contract under negotiation, a medical record, or a set of bank statements, that may be exactly what you were trying to avoid — the reasoning is set out in how browser-based PDF processing protects your privacy.
If you do share a link, set an expiry and restrict it to the recipient rather than to anyone holding the URL.
Three things that do not work
Zipping. This is the most common wasted step. A PDF’s contents are already compressed — the images inside a scan are stored in compressed form, and so are most content streams. Zipping compresses data that has already been compressed, and there is nothing left to remove.
Measured on a nine-megabyte image-based PDF:
| size | |
|---|---|
| the PDF | 9.27 MB |
| the same file zipped | 9.28 MB |
It got very slightly bigger. Zipping is useful for bundling several files into one attachment. It is not a size reduction.
Cropping. Cropping changes which part of a page is displayed; the content outside the crop is still in the file. Measured elsewhere, it adds around 140 bytes — see how to crop PDF margins.
Renaming, or changing the extension. The file is the size it is.
When the document is protected
One case that derails all of the above: an encrypted document cannot be compressed, split, or extracted from until it is opened with its password. Every operation needs to read the pages, and reading them requires decrypting them.
So the order is unlock, then reduce, then re-protect if the result still needs it. How to unlock a PDF when you know the password covers the first step, and it needs the correct password — there is no way around that and no honest tool offers one.
Worth noting that re-protecting at the end is not optional if the document was protected for a reason. A file you decrypted to make it smaller and then emailed unprotected has traded one problem for a worse one.
The order to try things
- Check the real limit — three quarters of the stated number.
- Select some text. Images or characters. This decides everything after it.
- Scan? Compress it.
- Text with images? Re-export from the source at a smaller setting.
- Do they need all of it? Extract the pages they asked about.
- Still too big? Split it, and name the parts clearly.
- Nothing else works? Share a link, with an expiry.
Stop at the first step that gets you under the limit.
If it is going to be sent regularly
A document you email once is a problem. A document you email monthly is a process worth fixing at the source.
Find where the size is being introduced — usually a scanner set to 600 DPI colour, or a template with an oversized logo on every page — and change it there. Every future copy is then small by default, and nobody has to remember a workaround.
Why any of this is possible
Compression is not bolted onto PDF; it is part of the format. ISO 32000-2 specifies the filters a compliant file may use for its streams and images, which is why a PDF arrives already compressed and why zipping it achieves nothing. The PDF Association publishes readable material on that structure.
It is also why the honest answer to “make this smaller” is so often send less of it rather than compress it harder. The format has already done the easy compression for you. What remains is a decision about content, and that decision is yours rather than the software’s.