That 45MB PDF won't attach to your email. Here's how to shrink PDF files by 80% while keeping text sharp and images clear — free, no software needed.
It's 4:47 PM on a Friday. You need to email a signed contract to your client before end of business. You click attach, pick the file, and Gmail politely tells you that your 38MB PDF exceeds the 25MB attachment limit. You try Outlook. Same problem — Outlook's limit is actually lower at 20MB. You briefly consider splitting the document across two emails and immediately realize how unprofessional that looks.
I've been there. More times than I care to admit.
The knee-jerk reaction is to Google "compress PDF free" and click the first result. You upload your confidential contract to some random website, wait for it to process on their server, download the result, and open it to find... blurry images, text that looks like it was photocopied five times, and a watermark asking you to upgrade to Premium.
There's a better way. And it doesn't require installing software, paying for subscriptions, or trusting a stranger's server with your documents.
This guide covers everything I've learned about PDF compression — why your files are huge in the first place, what "quality" actually means in a PDF context, and how to shrink files by 60-90% without the visual degradation that makes compressed PDFs look amateur.
Before we talk about compressing PDFs, it helps to understand why they get so big. A PDF isn't just "a document." It's a container format that can hold an absurd amount of data. Here's what's hiding inside that 45MB file you're trying to email.
When you create a PDF, the authoring software often embeds the entire font family used in the document. Not just the characters you used — the whole font. A single font file can be 200-500KB. If your document uses four fonts (a heading font, body font, bold variant, and italic variant), you're looking at 1-2MB just for fonts before any content is added.
Some design tools embed fonts multiple times if the same font appears in different elements. I've seen PDFs where the same font was embedded six times because it was used in the header, body, footer, tables, captions, and annotations separately.
This is the biggest culprit by far. Here's a scenario that happens thousands of times per day:
That photo you pasted into your report? It's displaying at 3 inches wide on the page, but the embedded image is 4032 x 3024 pixels — enough to print at 30 inches wide at 300 DPI. The PDF doesn't know you only need it to look decent at 3 inches. It keeps the whole thing.
A single uncompressed smartphone photo can be 8-15MB. Put five of those in a report and your "simple 10-page document" is suddenly 50MB+.
PDFs carry metadata you never see: author information, creation timestamps, modification history, application data, XML metadata streams, and sometimes even thumbnail previews of every page. In most cases this adds a few hundred KB, but documents that have been edited many times can accumulate significant metadata bloat.
Here's one most people don't know about: PDFs can contain duplicate internal objects. If you copy a page and paste it three times, some PDF editors duplicate every object on that page rather than referencing the original. The same image, the same font subset, the same vector graphic — repeated in the file's internal structure.
Fillable forms, JavaScript actions, embedded multimedia, annotations, bookmarks — every interactive element adds to file size. I once received a 90MB "PDF form" that turned out to contain an embedded video that auto-played when you opened it. Nobody asked for that.
If a PDF was exported from Illustrator, Photoshop, or InDesign with layers preserved, each layer increases file size substantially. A design proof PDF with 15 layers can be 5-10x larger than a flattened version.
Let me save you from the trial-and-error approach. Here are the current attachment limits for the services most people use:
| Email Provider | Attachment Limit | Notes |
|---|---|---|
| Gmail | 25 MB | Auto-converts to Drive link above this |
| Outlook / Microsoft 365 | 20 MB | OneDrive integration for larger files |
| Yahoo Mail | 25 MB | |
| Apple iCloud Mail | 20 MB | Mail Drop for up to 5 GB |
| ProtonMail | 25 MB | |
| Zoho Mail | 20 MB | |
| AOL Mail | 25 MB | |
| Corporate Exchange | 10-25 MB | Often set lower by IT admins |
Here's the thing people miss: these limits apply to the encoded email, not just your attachment. Email attachments are Base64-encoded, which increases the file size by roughly 33%. So that 25MB Gmail limit? Your actual attachment needs to be around 18-19MB to account for encoding overhead plus the email body itself.
Translation: if you want your PDF to reliably email through any provider, aim for under 15MB. Under 10MB is even safer if you're dealing with corporate email systems where IT departments set aggressive limits.
And it's not just email. Many web forms, job application portals, government submission systems, and insurance claim platforms cap uploads at 5-10MB. I once couldn't submit a health insurance claim because the portal maxed out at 4MB and my scanned receipts totaled 12MB.
Before we compress anything, let's define what we're protecting. When someone says "compress PDF without losing quality," what does "quality" mean?
Good news: text in a PDF is stored as vector data — mathematical descriptions of character shapes. You literally cannot degrade text quality through compression. Text takes up almost zero space and compresses with zero quality loss. If your PDF is text-only (a legal brief, a novel manuscript, code documentation), compression can shrink it dramatically with absolutely no visual change.
The exception is scanned documents where text is actually an image. We'll cover that scenario specifically.
This is where compression gets interesting and where most tools get it wrong.
Images in PDFs can be compressed in two ways:
Lossless compression reduces file size by finding redundant data patterns, like a zip file. The decompressed image is bit-for-bit identical to the original. You lose zero quality. The downside? The compression ratio is modest — typically 10-30% reduction.
Lossy compression achieves much higher compression ratios (50-90%+) by permanently removing image data that's deemed "less important" visually. This is where quality loss can happen. The question is whether that loss is perceptible.
Here's what matters: resolution appropriate for the viewing context. A 4000-pixel-wide photo embedded in a PDF that will only be viewed on screen at 800 pixels wide has 3200 pixels of data that provides zero benefit. Downsampling that image to 800 pixels is technically "lossy" — you've permanently removed data — but the visual result on screen is identical.
This is what smart compression tools do. They don't just crank down JPEG quality until your images look like they were compressed by a potato. They analyze how images are actually used in the document and optimize accordingly.
Charts, diagrams, logos, and illustrations created in vector format compress beautifully with zero quality impact. Vector data is already compact, and lossless compression techniques reduce it further without touching visual fidelity.
As mentioned, fonts compress losslessly. Better yet, smart compression tools can subset fonts — removing character definitions for letters not used in your document. If your PDF only uses the letters A-Z and 0-9 in a particular font, why keep definitions for Cyrillic, Greek, and mathematical symbols?
Font subsetting alone can save 200-400KB per font. In a document with multiple fonts, that's a meaningful reduction with literally zero change to how the document looks.
Most PDF compression tools offer some version of these three levels. Understanding what each does helps you pick the right one.
A 30MB document might drop to 21-27MB. Not dramatic, but sometimes enough to squeak under an email limit.
That 30MB document now drops to 9-18MB. This is the sweet spot for most people.
Your 30MB document could drop to 3-9MB. Huge savings but with visible trade-offs.
Let me walk through some actual scenarios with realistic file sizes, because abstract percentages don't help when you're staring at a 50MB file wondering if compression will actually save you.
Original: 47MB PDF (35-slide presentation with 22 smartphone photos)
Medium compression result: 8.2MB (82% reduction)
Why such a dramatic reduction? Because those smartphone photos were embedded at their original 12-megapixel resolution. The compression tool downsampled them to 200 DPI at their display size on the slide, which is more than enough for screen viewing and standard printing. Each photo went from ~3-5MB to ~200-400KB.
Quality verdict: Looked identical on screen. Printed on an office laser printer at letter size, I couldn't tell the difference. You'd only notice on a large-format print.
Original: 22MB PDF (15-page contract scanned at 600 DPI color)
Medium compression result: 3.8MB (83% reduction)
Scanned documents are the compression dream. The original was scanned at 600 DPI in full color, which is massive overkill for a text document. Compression reduced it to 200 DPI and converted from color to grayscale (since the document was just black text on white paper anyway).
Quality verdict: Text was perfectly legible. Actually looked slightly better because the grayscale conversion removed the slight yellowish tint from the scanner.
Original: 34MB PDF (8-page marketing brochure with professional graphics)
Low compression result: 28MB (18% reduction)
Medium compression result: 11MB (68% reduction)
Design documents are trickier because the images are often already optimized by the designer. Low compression only removed structural bloat. Medium compression downsampled images to 200 DPI which is fine for review purposes but the designer specifically asked me not to use the compressed version for print production.
Quality verdict: Fine for review and approval. Not suitable for final print production.
Original: 4.2MB PDF (120-page annual report, mostly text with a few charts)
Medium compression result: 1.8MB (57% reduction)
The savings came primarily from font subsetting and metadata removal. The few embedded charts were vector graphics that compressed losslessly.
Quality verdict: Pixel-identical. Zero visual difference whatsoever.
Original: 85MB PDF (24-page photography portfolio, full-bleed images)
Medium compression result: 19MB (78% reduction)
Low compression result: 62MB (27% reduction)
This is a case where you might want low compression instead of medium. The whole point of a photography portfolio is image quality. Medium compression still looked great on screen, but the photographer noticed a difference when pixel-peeping the images at 100% zoom.
Quality verdict: Medium was fine for emailing to a casual viewer. Low compression was better for sending to art directors or agencies who might scrutinize image quality.
Here's the straightforward process. I'll use the PDF compression tool on akousa.net as an example because it processes files in your browser (your document never gets uploaded to a server), but the general steps apply to any reputable tool.
Navigate to the PDF compress tool. You'll see a drop zone where you can drag and drop your file, or click to browse.
Drag your PDF into the tool or click to select it from your files. The tool will read the file and show you its current size.
Since processing happens in your browser, there's no upload step. The file goes from your hard drive to your browser's memory. This matters if you're compressing a confidential document — nothing leaves your machine.
Select the appropriate level based on your needs:
If you're not sure, start with medium. You can always re-compress with a different setting if the result doesn't meet your needs.
Click the compress button. Depending on the file size and complexity, this takes a few seconds to a minute. Since everything happens locally, speed depends on your device — a modern laptop handles a 50MB PDF in under 30 seconds.
Check the result. Good tools show you the before and after file size, plus the percentage reduction. Open the compressed file and spot-check a few pages — especially pages with images.
If you're happy, download the compressed file. If the quality isn't where you need it, go back and try a lighter compression setting.
That's it. Five steps. No account creation, no email verification, no "free trial" nags, no watermarks.
Sometimes you don't have one large PDF — you have twenty medium ones. Maybe you're a teacher compressing assignment handouts for your school's LMS. Maybe you're an accountant getting client tax documents under the portal's upload limit. Maybe you're just cleaning up a folder of scanned receipts.
For a small number of files, just run them through the compression tool one at a time. It takes about 30 seconds per file. Five files, maybe three minutes total. Not worth overthinking.
akousa.net's PDF tools support batch operations — you can drop multiple files at once and compress them all in a single operation. Each file gets compressed individually (so different files can contain different types of content and still get optimal compression), and you download the results as individual compressed files.
This is dramatically faster than processing files one by one, especially when you're dealing with 20+ documents.
A tip from experience: create a "compressed" subfolder before you start. When you download the compressed versions, save them to that folder. This avoids the nightmare of accidentally sending someone the original 40MB version because you grabbed the wrong file.
I name my compressed files with a suffix — report_compressed.pdf or contract_2026_sm.pdf — so I can tell at a glance which version is which.
The scenario: you're at a conference, someone hands you a printout, you scan it with your phone's camera app, and now you need to email that scan immediately. Your phone helpfully created a 25MB PDF from the scan. Your email app refuses to send it.
The built-in Files app doesn't have a PDF compression feature (as of 2026, Apple still hasn't added this). Your best options:
Browser-based tools: Open Safari and use an online compression tool that processes locally. akousa.net works in mobile Safari — drag and drop is replaced by a file picker, but the process is the same. The compression happens on your device.
Reduce scan quality at the source: If you're using the built-in document scanner (Notes app or Files app), scan in grayscale instead of color. This alone can reduce file size by 60-70% for text documents.
Shortcuts app: You can create an automation that uses the built-in PDF filter to reduce quality. It's clunky to set up but works once configured.
Similar options:
Browser-based tools: Chrome on Android handles in-browser PDF processing well. Same tools, same process.
Google Drive scan: If you use Google Drive's built-in scanner, it produces reasonably optimized PDFs by default. Usually smaller than what the stock camera app creates.
Adjust scan resolution: Most Android scanning apps let you choose DPI. For text documents, 200 DPI in grayscale is plenty. Don't default to 600 DPI color.
I keep using these terms, so let me explain them properly. This isn't academic — understanding the difference helps you make better decisions about which compression level to use.
Think of lossless compression like reorganizing a messy closet. Everything is still there — you've just arranged it more efficiently so it takes up less space. When you "decompress" (open the closet), everything comes back out exactly as it was.
In PDF terms, lossless compression:
The result is a smaller file that is visually and functionally identical to the original. No information is permanently deleted.
Typical reduction: 10-30%
Lossy compression is like editing a photo to remove distracting background clutter. The photo looks better for its intended purpose, but you can't get the clutter back. Some information is permanently gone.
In PDF terms, lossy compression:
The result is a smaller file where images may have slightly less detail than the originals, but the document serves its purpose effectively.
Typical reduction: 50-90%
Use lossless when:
Use lossy when:
For most everyday situations, lossy compression at a medium setting is perfectly fine. The quality loss is invisible to anyone who isn't pixel-peeping at 400% zoom.
Compression isn't always the right answer. Here are situations where you should think twice.
If a PDF has been digitally signed, notarized, or certified, compressing it will invalidate the signature. The digital signature is a cryptographic hash of the exact file contents. Change one byte and the signature breaks.
If you need to transmit a large signed document, use a file-sharing service (Google Drive, Dropbox, OneDrive) and send the link rather than compressing the file.
If a PDF is going to a professional print shop for offset printing, do not compress it. Print production requires:
Compressing a print-ready PDF can downsample images below print quality, convert CMYK to RGB, subset fonts in ways that cause substitution at the RIP, and strip production marks. Your printer will either reject the file or produce subpar output.
If you're creating a permanent archive of important documents, keep the uncompressed originals. Compress copies for daily use, but store the full-fidelity version somewhere safe. Storage is cheap — losing image data from your only copy of a document is not recoverable.
PDFs containing medical images (X-rays, MRIs, pathology slides) should never be compressed with lossy algorithms. Diagnostic information could be degraded. Medical imaging has its own specialized formats (DICOM) and compression standards for exactly this reason.
If you plan to re-open a PDF in an editor and make changes, work with the uncompressed version. Compressed PDFs are fine for viewing and sharing, but editing a compressed file and then saving it can compound quality loss — similar to repeatedly re-saving a JPEG.
Here's the approach that power users take: optimize your content before creating the PDF in the first place.
This is the single most impactful thing you can do. Before dropping photos into your Word document, PowerPoint, or design tool:
Resize images to the size they'll display. If an image will be 6 inches wide in your document, resize it to 1800 pixels wide (300 DPI) or 1200 pixels wide (200 DPI). Don't insert a 4000-pixel-wide image that will display at 6 inches.
Compress images individually. Run your photos through an image compression tool first. akousa.net has image compression tools that can reduce a 5MB JPEG to 800KB with negligible visual change. Do this before inserting into your document.
Use the right format for the right content. Photos should be JPEG. Screenshots with text should be PNG. Simple graphics with few colors should be PNG or SVG. Don't paste a 15MB PNG screenshot when a 200KB JPEG would look identical.
Crop before inserting. Don't insert a wide landscape photo and then use your document editor's crop tool. Crop the image in a photo editor first. Many document editors preserve the original uncropped image data — your PDF contains the full image even though only part of it is visible.
Most applications that export to PDF give you options:
Microsoft Word / PowerPoint:
Google Docs / Slides:
Adobe InDesign / Illustrator:
macOS Preview:
Sounds obvious, but: if you're emailing a 200-page report and the recipient only needs pages 45-60, extract those pages and send only what's needed. A 200-page document at 50MB becomes 16 pages at maybe 4MB.
akousa.net has a PDF split tool that lets you extract specific page ranges — useful for exactly this scenario.
If your PDF has fillable form fields that are already filled in, flattening them (converting interactive fields to static content) can reduce file size. Same with annotations, comments, and markup — if the review process is complete, flatten everything into the page content.
Scanned PDFs are special because every page is essentially a photograph. A 20-page scan at 300 DPI color can easily be 40-60MB. Here's how to get those down dramatically:
Convert to grayscale if the original document is black and white. This alone cuts file size by 60-70%.
Reduce DPI to 200 for documents that will be read on screen. If the text is standard size (10-12 point), 200 DPI is perfectly legible.
Consider OCR first. Running OCR (optical character recognition) on a scanned document creates a text layer. Some tools then allow you to reduce the image quality substantially because the text is preserved in the text layer — you get tiny files that are still perfectly readable and searchable.
Use medium compression. For scanned documents, medium compression gives the best ratio of size reduction to quality. Light compression doesn't do enough; strong compression can make text fuzzy.
If you regularly create the same type of document (monthly reports, weekly newsletters, quarterly reviews), invest the time once to optimize your template:
A well-optimized template produces PDFs that are 30-50% smaller than the default, before any post-creation compression. Combined with compression after export, you're looking at files that are 60-80% smaller than what you'd get with default settings.
I've made all of these. Learn from my mistakes instead.
If a PDF has already been compressed, running it through compression again produces diminishing returns at best and quality degradation at worst. Each round of lossy compression removes more data. After two or three rounds, the quality loss becomes visible — especially in photographs and detailed graphics.
Solution: Always compress from the original, uncompressed source. If you need a smaller file than your current compressed version, go back to the original and compress more aggressively in one pass.
"Why not always use maximum compression?" Because it produces visibly degraded output for image-heavy documents. Maximum compression is designed for situations where readability matters but visual fidelity doesn't.
Solution: Start with medium compression. Only go higher if you need to, and check the result before sending.
Don't compress a PDF and then realize you need to add another page, which means going back to the source, adding the page, re-exporting, and re-compressing. Compression should be the last step in your workflow.
Solution: Finalize your document completely, then compress once as the final step before distribution.
I once compressed a 50-page report and emailed it without opening the compressed version. The compression tool I was using at the time had converted all my charts to low-resolution blurs. The client noticed.
Solution: Always open the compressed file and spot-check at least a few pages before sending. Pay attention to images, charts, and any fine detail.
This is the big one. Most online PDF tools upload your file to their servers for processing. Your medical records, tax returns, employment contracts, financial statements — sitting on a server you know nothing about.
Solution: Use tools that process files locally in your browser. The document never leaves your device. You can verify this by disconnecting from the internet and seeing if the tool still works (browser-based tools will continue to function; server-dependent tools won't).
Here's a cheat sheet you can reference whenever you need to compress a PDF:
| Document Type | Recommended Level | Expected Reduction | Notes |
|---|---|---|---|
| Text-only report | Light | 20-40% | Lossless is sufficient |
| Business report with charts | Medium | 40-60% | Charts compress well |
| Photo-heavy presentation | Medium | 60-85% | Biggest gains here |
| Scanned document | Medium | 70-85% | Convert to grayscale if B&W |
| Design proof | Light | 15-25% | Preserve image quality |
| Photography portfolio | Light or skip | 15-30% | Quality is the point |
| Tax forms / legal | Light only | 15-25% | Don't risk quality issues |
| Filled-out form | Medium | 40-60% | Flatten fields first |
| Current Size | Target: Under 20MB | Target: Under 10MB | Target: Under 5MB |
|---|---|---|---|
| 25 MB | Light compression | Medium compression | Medium-high compression |
| 50 MB | Medium compression | Medium-high compression | High compression + check quality |
| 100 MB | Medium compression | High compression | May need to split document |
| 200+ MB | Medium compression | Split into sections + compress | Definitely split first |
I want to spend a moment on this because it's important and most people don't think about it.
When you compress a PDF using a server-based tool, your document travels across the internet to someone else's computer, gets processed there, and then travels back to you. During this journey:
For a recipe collection or a public flyer, this is fine. For a signed employment contract, a medical report, your tax return, or confidential business documents? You're trusting a lot of infrastructure and a lot of people with sensitive information.
Browser-based PDF tools that process locally are fundamentally different. Your file goes from your storage to your browser's memory, gets processed by code running on your device, and the result is saved back to your storage. The document never touches a network. There's nothing to intercept, nothing to breach, nothing to subpoena.
akousa.net's entire PDF suite works this way — all 56 tools process documents locally in your browser. It's not just the compression tool. Merging, splitting, converting, signing, watermarking, OCR — all local.
This isn't a sales pitch. It's a practical consideration that matters when you're handling documents that contain personal or confidential information.
Sometimes compression isn't the best approach. Here are other ways to deal with large PDFs.
Instead of attaching a 50MB PDF to an email, upload it to Google Drive, Dropbox, or OneDrive, and share a link. Zero quality loss, works for any file size, and the recipient always gets the latest version. The downside: some people are suspicious of links in emails, and links can expire.
If your document is large but the recipient only needs specific sections, extract those pages. A 100-page document split into a 15-page excerpt is going to be much smaller regardless of compression.
When this works best: Reports where the recipient needs specific chapters, multi-project proposals where each client only needs their section, meeting minutes where attendees need their department's portion.
If you create the same type of document regularly, use lower-resolution images from the start, export with optimized PDF settings, and use vector graphics instead of rasterized charts. Prevention is easier than cure.
No. Text data in a PDF is preserved through compression. You can still search, copy, and select text in a compressed PDF. The exception is if you compress a scanned document so aggressively that the OCR layer is affected — but any decent compression tool handles this properly.
You'll typically need to enter the password first. Most compression tools can process password-protected PDFs if you provide the password. The compressed output can be re-encrypted if needed.
You can, but you shouldn't compress more than once with lossy settings. Each round of lossy compression degrades quality further. If you need more compression, go back to the original and compress more aggressively in a single pass.
Reputable compression tools preserve bookmarks, table of contents, internal links, and document structure. Only aggressive optimization tools that strip "all metadata" might remove these. Always check after compression.
PDFs under 1MB generally don't benefit much from compression. The overhead of the PDF structure itself is a larger percentage of the file, and there's less redundant data to eliminate. If your PDF is already under 1-2MB, it's probably fine as-is.
No. Lossy compression permanently removes data. You cannot "uncompress" a PDF back to its original quality. This is why you should always keep the original uncompressed file and compress copies for distribution.
Yes. Any modification to a digitally signed PDF — including compression — invalidates the signature. If a PDF is signed, share it at its current size via a file link rather than compressing it.
Yes, but less dramatically. For text-heavy multi-page documents, compression mainly benefits from font subsetting and structural optimization. You might see 20-40% reduction. If you need more, consider whether all those pages need to be in a single file.
PDF compression isn't complicated, but it's easy to get wrong. The key takeaways:
Understand why your PDF is large before choosing a compression method. Image-heavy documents respond differently than text-heavy ones.
Medium compression is right 80% of the time. Start there. Only go lighter for legal/print documents or heavier when you're desperate.
Compress from the original, once. Don't re-compress compressed files.
Always check the result before sending. Ten seconds of previewing saves you from the embarrassment of sending a document with blurry images.
Use local processing tools for anything confidential. Your medical records shouldn't visit someone else's server just to get smaller.
Optimize before you PDF when possible. Right-sized images and proper export settings mean less compression needed after the fact.
Know when not to compress. Signed documents, print-ready files, and archival masters should stay at full fidelity.
The days of struggling with PDF file sizes are over. Between browser-based compression tools and the strategies in this guide, you can get virtually any PDF to any target size without making it look like it was faxed in 1997.
Go compress that file and catch your 5 PM deadline. The tool is free, it runs in your browser, and your document stays on your machine. No excuses.
Akousa