How to Compress a PDF Without Losing Quality

You need to email a PDF and your provider has a 10 MB limit. The file is 32 MB. This is where every guide on the internet suddenly wants you to upload your file to a random server. This one doesn’t.

The honest version: PDF compression comes in two flavours, they work very differently, and both have tradeoffs you should know before clicking a button.

Why PDFs get huge in the first place

A PDF is a container. Inside can be text, vector graphics, embedded fonts, and images. Text and vectors are tiny — a 100-page legal contract that’s all text is under 1 MB. Images are what push file size up.

If your PDF is over 10 MB, the odds are 90%+ that images are the reason. A scanned document at 300 DPI, a report with high-res charts, a portfolio full of photos — those are 10-50 MB files easily.

The two ways to compress

1. Structural (lossless) compression

The PDF format has room for cruft — duplicated objects, unused fonts, uncompressed streams that could be zip-compressed. A structural compressor cleans this up without changing any actual content. Text stays crisp. Images stay pristine.

Savings you can expect: 5-20% for a typical PDF. On a hand-crafted PDF from a decent tool, sometimes nothing at all — no cruft to remove.

Use it when: the PDF was generated by an application (Word, Google Docs, InDesign) rather than scanned from paper. Also always safe as a first pass because you can’t lose quality.

2. Image re-encoding (lossy) compression

This is the technique that actually shrinks big files. The compressor rasterizes each page (turns it back into a picture) at a chosen resolution, re-encodes it as a JPEG at a chosen quality, and rebuilds a new PDF from those images. This is what our PDF Compress tool does.

Savings you can expect: 50-90% on scanned or image-heavy PDFs. A 30 MB scan can end up at 3-5 MB with quality still perfectly readable.

The tradeoff: everything becomes an image. Text is no longer selectable or searchable — it’s pixels. If someone needs to copy-paste text out of the file, or if it needs to work with a screen reader, this method breaks that.

Quality settings — what to pick

• High quality — scale ≈ 2.0×, JPEG quality 85. Barely visible difference from the original. Modest size reduction (typically 30-50%). Good for archival or documents you might print.
• Medium quality — scale ≈ 1.5×, JPEG quality 70. Slight quality drop noticeable if you zoom in but invisible at normal reading size. Substantial size reduction (60-80%). Good default for email.
• Low quality — scale ≈ 1.0×, JPEG quality 50. Blocky under close inspection, fine at reading distance. Massive reduction (80-95%). Use when the file MUST fit under a strict size cap.

Things that DON’T shrink a PDF meaningfully

• Renaming the file. Obvious but people try it.
• Zipping it. PDFs are already internally compressed. Adding a zip wrapper usually saves <1%.
• “Save as reduced size” in Preview / Acrobat. This uses lossy image re-encoding under the hood — the same technique as any online compressor, just with a nicer button. Fine to use; there’s no secret sauce.
• Printing to PDF from the current PDF. Ironically sometimes makes the file bigger. The re-rendering process doesn’t inherit the original’s image compression settings.

Practical decision tree

• Text-heavy document from Word or Google Docs, currently >5 MB? Something odd is going on — probably embedded fonts or images. Try structural compression first. If that doesn’t help, look for embedded images to remove or link externally.
• Scan of a paper document, currently >10 MB? Lossy compression at medium quality. You’ll likely go from 30 MB to 3-5 MB with the text still perfectly legible.
• Portfolio or report with high-res photos, need to email? Lossy compression at medium quality. Also consider whether every page needs to be at print resolution or if screen resolution is fine.
• PDF needs to remain searchable / accessible? Structural only. Accept a modest size reduction. Alternative: extract the images, compress them separately with our Image Resizer, then embed them back with a tool that preserves the text layer (needs a dedicated PDF editor).
• Need to combine several PDFs into one instead? PDF Merge does that without recompressing anything.
• Have a bunch of image files you want as one PDF? Images to PDF combines them with page-size and fit-mode options.