How to Compress a PDF Without Losing Quality
To compress a PDF without losing quality, use Optimize PDF for text documents and Compress PDF for scans. Optimisation re-packs the file losslessly — nothing changes visually. Compression re-encodes the images inside, which is where the large savings are. Both run in your browser; your document is never uploaded.
A 40 MB file that will not attach to an email is one of the most common document problems there is. This guide covers why it happens, which method actually applies to your file, and what to do when you need to hit a specific size.
Why is my PDF so large?
Almost always images — usually because the pages are photographs of paper rather than text. A scanned page at 300 DPI is a 2,550 × 3,300 pixel image, and thirty of those make a large file regardless of what the words say.
The four common causes, in rough order of frequency:
- Scanned pages. Every page is an image. This is the single biggest cause and also the most compressible.
- Embedded images at full resolution. A photo placed small on the page is often still stored at its original size.
- Embedded fonts. Usually a few hundred kilobytes, occasionally several megabytes for CJK or icon fonts.
- Incremental save history. Each save can append rather than rewrite, so a heavily edited file carries earlier versions inside it.
If you are not sure which applies, run the file through PDF Size Analyzer. It reports each page's byte contribution, and a single 8 MB page in a 10 MB document points straight at the culprit.
What does compressing a PDF actually do?
It re-encodes the images inside the document at a lower quality setting. It does not touch the text. This is the detail that explains almost every surprising result people get.
Because compression works on images, the outcome depends entirely on what your PDF contains:
| Document type | Typical reduction |
|---|---|
| Scanned pages (images) | 60–80% |
| Photo or chart-heavy report | 40–70% |
| Mixed text and images | 20–50% |
| Text-only (contract, ebook) | 0–10% |
A text-only PDF that barely changes is not a failure of the tool. There was nothing image-shaped to compress.
Which method should I use — optimise or compress?
Use Optimize PDF when quality must be untouched, and Compress PDF when the file must be smaller.
Optimize PDF is lossless. It removes orphaned objects, duplicated resources and incremental-save history, then re-packs the structure. No image is re-encoded and no pixel changes. Savings vary enormously — a heavily edited file can shrink substantially while a freshly exported one barely moves.
Compress PDF is lossy for images. Four levels are offered and the resulting size is shown before you download, so you can step down one level at a time rather than guessing.
A practical sequence: optimise first, check the size, and only compress if you still need to be smaller. Many files are solved by the lossless step alone.
Does compressing a PDF reduce quality?
Text stays sharp at every level; images lose detail progressively. Compression never rasterises your document, so text remains selectable, searchable and crisp on screen and in print.
What degrades is photographic detail. At Low and Medium the difference is usually invisible on screen. At Maximum, scanned text can start to show softening around the letterforms — still readable, but visibly processed if you compare side by side.
Two things worth knowing:
- Compression is not reversible from the output file. Keep the original until you have checked the result.
- It is cumulative. Compressing an already-compressed PDF degrades it again. Always start from the source document.
How do I get a PDF under a specific size?
Work down the levels and check the reported size before downloading — the tool shows you the result of each setting rather than making you download to find out.
The limits people are usually fighting:
- Gmail and Outlook: 25 MB per message
- Many corporate mail servers: 10 MB
- Government and university portals: frequently 2 MB or 5 MB per document
- Some visa and job application forms: as low as 500 KB
If compression alone will not get you there, the file is probably the wrong shape for the constraint. Splitting is often the better answer: Split PDF can divide a document into parts that each fit under the cap, which is usually more acceptable to a portal than a badly degraded single file.
Why did compressing barely change my file?
Because there were no images to compress. If your PDF is a contract, an ebook or an exported text report, most of its bytes are fonts and structure, and image compression has nothing to work on.
For those files, try Optimize PDF instead. If it is still large, the cause is usually embedded font subsets or accumulated save history, and lossless re-packing is the only thing that will help.
There is one more case worth checking: a document that looks like text but was scanned. If you cannot select the words with your cursor, it is an image, and it will compress well.
Should I compress before or after OCR?
Run OCR first, then compress. Text recognition reads the page images, so recognition accuracy depends on image quality — compressing first throws away detail the OCR engine needs.
The usual order is: scan → OCR → compress. That gives you a searchable document at the smallest workable size, rather than a small document that OCR then struggles to read.
Is it safe to compress a confidential PDF online?
It depends entirely on whether the tool uploads your file — and most do. In a transport-layer measurement of 20 online PDF tools published on 11 August 2026, 12 of the 13 services that could be measured transmitted the entire test file to their own servers. Filevane transmitted 0.00 MB.
That matters more for compression than for almost any other task, because the documents people compress are the ones that are too big to email: scanned contracts, medical records, tax returns, ID documents.
You do not have to take our word for it. Open your browser's Network tab, run the tool, and watch for an outbound request carrying your file — there is not one. Our privacy proof page publishes the Content-Security-Policy header that makes an upload impossible rather than merely against policy.
Compress a PDF now
Compress PDF — four quality levels, size shown before download, no account, no watermark, no file-size cap beyond your own device's memory. Your document never leaves your browser.
Working with a text document instead? Start with Optimize PDF for a lossless reduction. Not sure what is making the file large? PDF Size Analyzer will tell you in a few seconds.
Tools used in this guide
Frequently asked questions
How do I compress a PDF without losing quality?
Use Optimize PDF, which is lossless: it re-packs the file's internal structure, removes orphaned objects and duplicated resources, and re-encodes nothing. Text and images are pixel-identical afterwards. If you need a larger reduction than that achieves, Compress PDF re-encodes the images inside the document — text still stays sharp and selectable at every level, but photographic detail is reduced.
Why is my PDF so large?
Usually because the pages are images rather than text. A scanned page at 300 DPI is roughly a 2,550 x 3,300 pixel image, and a document made of thirty of those is large regardless of its word count. The other common causes are embedded images stored at full resolution despite appearing small on the page, embedded font subsets, and incremental save history left behind by repeated editing.
How much smaller will my PDF get?
It depends almost entirely on whether the document contains images. Scanned pages typically reduce by 60-80%, photo-heavy reports by 40-70%, mixed documents by 20-50%, and text-only PDFs by 0-10%. A text document that barely changes is not a tool failure — compression works on images, and there were none to work on.
Does compressing a PDF make the text blurry?
No. Compression re-encodes the images inside the document and never rasterises the page, so text remains selectable, searchable and sharp at every level. What degrades is photographic detail. At Maximum quality on a scanned document, letterforms can start to soften slightly, because on a scan the text is itself part of an image.
How do I get a PDF under 2MB for an upload form?
Step down the compression levels and check the reported size before downloading. If compression alone will not reach the cap, splitting the document into parts that each fit is usually a better answer than degrading a single file badly — most portals accept multiple uploads. Removing unnecessary pages first also shrinks a file far more reliably than compression does.
Should I compress before or after running OCR?
Run OCR first, then compress. Text recognition works from the page images, so accuracy depends on image quality, and compressing first discards detail the engine needs to read the characters. The reliable order is scan, then OCR, then compress.
Is it safe to compress a confidential PDF with an online tool?
Only if the tool does not upload it, and most do. A transport-layer measurement of 20 online PDF tools published on 11 August 2026 found that 12 of the 13 services that could be measured sent the entire test file to their own servers. Filevane transmitted 0.00 MB. This matters especially for compression, because the documents people need to shrink are usually the sensitive ones — scanned contracts, medical records and tax returns.