How to Compress a PDF for Email (Windows, Mac, Phone and Browser)

The short answer

To compress a PDF for an email attachment: open the file in a compressor that rebuilds it from page images, choose a smaller preset, and download the result. If the document is a scan, a photo export, or something full of pictures, this drops the size fast and the pages still read cleanly on screen.

If the document was created as text — a Word export, a generated report, a form — the same move can backfire. Rebuilding from images throws away the searchable text layer, the links and the form fields, and on a text-heavy file the new image-based PDF is often larger than the original you started with. That is not a bug in any particular tool; it is what the method does.

So decide before you click anything: is this a pile of pictures that happens to be in a PDF, or is it real text you want the reader to be able to search and copy? The first case is exactly what image rebuilding shrinks. The second case is where you should read the two sections below before you compress anything, because the wrong choice costs you content you cannot get back without the original file.

• Scanned or photo-heavy PDF: image-based compression is the fastest real win.
• Text-heavy PDF: compression here can enlarge the file and kill search, links and forms.
• Attachment limit is about the finished file size, so aim to land under it with room to spare.
• One file at a time, three presets, no account, and the document never leaves your machine.

How this compression works, and what it costs

There are two honest ways to make a PDF smaller, and they behave completely differently. The first is lossless optimisation: the tool keeps every object in the document and only removes redundancy it can prove is safe to remove — duplicate embedded fonts, unused data, inefficient encoding. The text stays text, links stay links, and a searchable document stays searchable, but the size drop is modest on a file that was already built well.

This tool does the second thing, and it is important not to confuse them: it rebuilds your PDF from page images. Under the hood it opens the document, renders every page to a flat picture, re-encodes each of those pages as a compressed JPEG, and draws the images into a brand-new PDF. Your original document is not optimised — it is effectively screenshotted, page by page, and reassembled.

That is why the size comes down so hard on a scan or a photo export. Those files store enormous image data, and re-encoding the page as a tighter JPEG is exactly the lever that matters. Each page is flattened onto a white background as part of the rebuild, which is fine for ordinary documents and is how transparency and layered content get resolved into a single picture.

The cost is real and you cannot get it back from the output: the text layer is gone, so nobody can select, search, or copy text from the compressed file; hyperlinks stop working; and interactive form fields become flat pictures of fields. This tool has no text recognition step to restore any of that — once a page is an image, it stays an image. For a scanned receipt you are emailing as proof, none of that matters. For a form you want someone to fill in, or a report someone will search, it matters completely.

What the text-heavy trap looks like

The rebuild explains the one failure that surprises people. A page of clean text is already a tiny amount of data — the letters are stored as instructions, not as pictures. When you force that page through the image route, the tool replaces a few kilobytes of text with a full picture of the page, and a picture of a page is far bigger. Do that across a text-dense document and the output PDF can come out larger than the input.

The tool tells you when this has happened rather than letting you find out in the mail client: if the result is bigger than the original, it says so on screen, because that outcome is common for text-heavy PDFs. When you see it, stop and pick a different approach for that file.

So the decision rule is simple. Reach for image-based compression when the PDF is a scan, a photograph, a screenshot export, or anything where a page is mostly pixels. Leave text-based PDFs alone, or make them smaller some other way — by removing embedded fonts you do not need, downsampling the images that are actually inside them, or, if the real problem is just the message, pasting the content into the email instead of attaching the file.

What “email size” actually means

Before compressing, know what you are compressing toward. Attachment limits are set by the mail system, not by PDF, and many providers cap the total message rather than the file: your attachment shares the allowance with the message body, signatures, images in the signature, and anything else in that one email. On top of that, most mail systems expand a binary attachment as they encode it for transport, so a file that is just under the raw limit can still tip a message over it once encoding overhead is counted.

There is no single universal number to hit, and the caps differ between free consumer accounts and locked-down corporate gateways, so treat the size you are aiming at as comfortably under whatever your provider mentions, not right at the edge. A PDF that shrinks to well under the ceiling is what actually gets through on the first try.

This is the honest reason “compress for free” and “compress to a specific size” keep coming up as separate needs: the free browser route is not a service that promises a target file size, it is a tool with three quality presets that rebuilds your document. It shrinks a scan dramatically, and “enough for email” is usually easy to reach; landing on a precise byte target is not something these presets offer.

• The limit is on the whole message, not only on your PDF.
• Encoding overhead means a file can push the message over even when the file alone looks compliant.
• Aim well under the stated cap so corporate relays and reply chains still deliver it.

Choosing between the three presets

The tool gives you three compression levels as radio choices, not a slider: High, Balanced, and Small. There is no custom quality control and no field where you type a target size, which keeps the decision to picking one of three. That is a deliberate simplification, and it is also the honest one: presets can be re-run instantly, whereas a free-form size target on an unknown document is a promise a browser tool cannot reliably keep.

Read them as directions rather than numbers. High keeps the most detail from each page and shrinks the least; it is what you want for something with fine print, small type, or detail you need the reader to see. Small squeezes hardest and is what you reach for when a large scan simply has to fit an inbox. Balanced sits in the middle and is the sensible first try when you are not sure.

The practical workflow is to compress, look at the result, and if it is still too big for the attachment limit, run the next smaller preset. The tool has a “try another preset” action that re-runs on the file you already loaded, so you do not have to upload again to test the tighter setting. Compare the on-screen size for each output and stop at the first one that both reads acceptably and lands under your limit.

Step by step on each device

The good news for the “on mac” and cross-platform versions of this question is that the route is the same everywhere, because the work happens in the browser rather than in the operating system. Where devices differ is only in how you get the file and where the download lands.

The general path on all of them: open the compressor, add your single PDF, pick a preset, run it, and download the new file, which is saved with “-compressed” added to the original name. Then attach that output to your email, not the original.

On Windows

Open the tool in any modern browser — Edge, Chrome, or Firefox. Use the file picker to select one PDF from File Explorer or your mail client's download folder. Pick Balanced if you are unsure, or Small for a big scan, then run the compression. When the finished file appears, download it; your browser drops it in the Downloads folder. Attach the downloaded copy to the outgoing message and check its size in File Explorer before sending if you want to be certain it is under the limit.

On Mac

macOS has no dedicated PDF-compression button most people can find, which is exactly why “compress a PDF on mac” is such a common search, but the browser route here is faster than hunting for a system feature. Open the tool in Safari or Chrome, load one PDF, choose a preset, and run it. The download lands in Downloads by default. If you would rather work from Files, the browser hands you a normal file you can drag into a new Mail attachment from Finder afterward.

One Mac-specific note: if the PDF you are trying to shrink is text-heavy and the output comes out larger, the image method is the wrong tool for that particular file, and no amount of preset-changing fixes it. Leave that document as it is.

On iPhone and iPad

Open the tool in Safari. Add your PDF — if it lives in the Files app or in a Mail attachment, save it to Files first so the picker can reach it. Pick a preset, run the compression, and use the download to save the smaller PDF back into Files. From there you can attach it to a new email the way you attach any file. A phone is a browser like any other for this, so you are not installing an app to do it.

The one phone caveat is memory, not permission: a very large PDF from your camera roll can strain the tab more than it would a laptop, so if a huge file stalls, start with a moderate one and see how the device behaves.

On Android

Open the tool in Chrome. Load one PDF from your Files app or from a downloaded attachment, choose a preset, and run it. Download the result, which goes to your device's Downloads. Then compose the email and attach the smaller file. The steps are the same as on every other platform, because there is nothing here that depends on Android specifically.

The browser route, and why nothing is uploaded

This is not a front end that ships your document to a server and returns it smaller. The decompression and rebuild run inside the page on your own device: your browser renders the pages, re-encodes them, and assembles the new PDF in memory, then hands you the finished file as a download. The document's contents are never transmitted anywhere, which is why the same tool works just as well on a scanned contract or a medical page as it does on a landscape photo, with no upload step and no account.

One point people get wrong, so it is worth stating plainly: the processing engine that reads your PDF is loaded from a content-delivery network, so the very first time you open the tool you need an internet connection for the page itself to load. After that, your document is handled locally on your machine. “The file never leaves your device” and “the page needed a connection to open” are both true at the same time, and neither one contradicts the other.

Compressing a large PDF for email

A genuinely large document is where image-based compression earns its place, because size usually scales with how much image data is in there. A photo dump, a multi-page scan from a flatbed or a phone camera, a slide deck exported to PDF — these have enormous headroom, and the Small preset can take a file that a mail system rejects down to something that goes straight through.

There is no hard cap on the file size this runs on, but the practical limit is your own device: the pages are decoded and rebuilt in memory, so a very large or very long PDF is constrained by how much room your browser has to work in rather than by a queue or a quota. On a laptop you have more of that room than on a phone.

And it is one file at a time. The tool reads a single PDF per run, so several large documents to send means several runs. There is no batch queue. If your real job is “shrink these twelve scans,” the honest approach is to compress each one you need to send now, rather than expecting a bulk operation the tool does not offer. The upside of one-at-a-time is that you can eyeball every output before attaching it, which is exactly what you want when a wrong size bounces a message back to you.

For password-protected files there is one hard boundary: a protected PDF is rejected, because the tool cannot decrypt and render its pages. Remove the password on the original, then compress the unprotected copy.

Should you just zip the PDF instead?

“How to zip a PDF for email” is a common reflex, so it deserves a straight answer: zipping almost never solves an over-limit attachment, and it usually makes the email worse to receive.

A ZIP is a container, not a compressor for this purpose. It wraps files in an archive, and the compression it applies pays off on things that shrink well. A PDF is already a compressed format — its images and streams are encoded — so putting a single PDF inside a ZIP typically saves very little, and a scanned PDF that is already tightly packed barely gets smaller at all. You have added a step without removing the bytes that tripped the size limit.

The bigger problem is that most mail systems count the total against the size cap as though the archive were already opened — they look at the content, not just the wrapper — so a zipped PDF that expands to the same size you were trying to escape still lands over the limit once a recipient's server unzips and evaluates it. Some gateways simply block archives outright as a security measure, and a corporate filter that refuses ZIP attachments will bounce your message regardless of what is inside.

There is also the human cost. A recipient cannot preview a zipped PDF: they have to download it, open the archive, and extract the document before reading a word of it, which on a phone is a genuine nuisance and on a work machine can raise an alert. On one document for one reader, that friction buys nothing.

Zipping earns its place in a different situation: bundling several separate PDFs into a single download so one person gets a whole set. In that case the archive is for tidiness and delivery, not for size — and if the set is still over the limit, compress each PDF first, then bundle the smaller files. But when the task is “make this one PDF fit an attachment,” compressing the PDF is the better answer, not wrapping it.

Limits worth knowing before you start

Everything here is honest and small, but better known upfront than discovered mid-send. The tool takes one file at a time and runs three fixed presets, so there is no batch mode and no custom quality slider or target-size field — you pick High, Balanced, or Small and re-run with another preset if the result is not right.

Because it rebuilds from images, the output always loses the text layer, working links and interactive form fields, and that is not switchable. If any of those matter to the reader, this is the wrong tool for that file regardless of how much the size drops.

The engine that reads PDFs comes from a CDN, so opening the tool for the first time needs a connection even though your document is never transmitted once it loads. A protected PDF cannot be processed. And, as covered above, very large files are bounded by browser memory, so an unusually heavy document can struggle on a phone that handles it fine on a desktop.

None of that touches the core promise: the file you get back is smaller for image-heavy documents, the whole job runs on your device, and it costs nothing and asks for no account. Just pick the right document for it.

• One PDF per run — no batch or multi-file mode.
• Three presets — no custom slider and no target-size input.
• Always loses searchable text, links and form fields — no way to keep them.
• Password-protected PDFs are rejected — remove protection first.
• First load needs a connection because the reader engine comes from a CDN; your file itself is never uploaded.

A quick checklist before you hit send

Run this and you will almost never have a bounced attachment or a stripped document to deal with.

Know which PDF you have. Scan or photos — image compression is the right call. Real text — leave it alone or find another way, because this route can make it larger and will erase its searchable text.

Know what you are compressing toward. The limit is on the whole message, not the file, and the file's size counts after encoding overhead, so land comfortably under the cap your mail system mentions.

Pick Balanced for an uncertain file, Small for a big scan, High when fine detail must survive, then use “try another preset” rather than re-uploading if the first pass is still too big.

Check the output on screen: the new size, and whether the pages still read. If the tool reports the result came out larger, you have a text-heavy file — stop and reconsider.

Do not reach for a ZIP to beat a size limit; compress the PDF instead, and save zipping for bundling several already-small documents. Attach the compressed file, not the original, and you are done.

Frequently asked questions