How to reduce PDF file size without losing quality
Most of a PDF's size comes from its images, not its text — which is why the right fix is almost never 'compress harder' but 'compress the right thing.'
Why PDFs get bloated in the first place
Open a 40MB PDF and the text inside it — even a hundred pages of it — is rarely the problem. Text is stored as compact instructions ("draw this character at this position"), and a whole novel's worth barely registers. Images are a different story: a single full-page photo scanned at high resolution can be tens of megabytes on its own, and a report with a dozen embedded charts or product photos adds up fast.
Scanned documents are the extreme case. A scanner set to 600 DPI in colour produces images many times larger than the same page would need for on-screen reading, because scanner defaults are tuned for archival fidelity, not file size. Multiply that by every page and a 20-page scanned contract can easily outweigh a 200-page text-only report.
The rest — embedded fonts, unused metadata, redundant object streams left behind by repeated edits — matters too, but usually accounts for a rounding error next to what images cost.
What actually shrinks a PDF
Two things move the needle, and everything else is a footnote:
- Re-encoding images at a lower quality setting — the same idea as saving a JPEG at 70% instead of 100%. Past a certain point this is invisible on screen and only marginally noticeable in print.
- Downsampling — reducing an image's actual pixel dimensions to match what the page will realistically display it at. A photo scanned at 600 DPI but shown at half a page doesn't need anywhere near that resolution to look sharp.
- Stripping embedded fonts you don't need, unused thumbnails, and edit history metadata that accumulates from repeated saves in some editors.
How much quality loss is actually visible
This is the part most compression tools don't explain, and it's the difference between a PDF that looks fine and one that looks smudged.
For anything read on a screen, 150 DPI is the practical ceiling of what a human eye distinguishes from 300 DPI at normal reading distance — screens themselves rarely exceed that effective density for a full page. For print, 300 DPI is the standard target; going lower starts to show at the printed edge of text and fine lines.
On JPEG quality specifically, the range between 100% and about 80% is where compression is doing the most work for the least visible cost — most of the file-size reduction happens early, while the visible artefacts (blockiness around sharp edges, banding in smooth gradients) don't show up until you drop well below that.
In practice: start around 150 DPI and 80% JPEG quality for a document that's mainly going to be read on screen, and go up on both for anything meant to be printed or that contains fine detail like architectural drawings or line art.
When compression won't help
If a PDF is already made of efficiently compressed images — say, it was exported from a design tool that already optimised them — running it through another compression pass buys very little, since there isn't much slack left to remove.
And if the actual goal is a smaller, searchable, professional-looking document from a stack of scans, compression is the wrong tool entirely. What that needs is OCR to add a text layer, which addresses the real complaint ("this file is unwieldy") more completely than shrinking the images ever will.
Frequently asked questions
Will compressing a PDF make the text blurry?
No — compression here targets images specifically. Text stored as real vector instructions (rather than a scanned image of text) stays crisp regardless of compression level.
What's a safe compression level to start with?
Around 150 DPI and 80% JPEG quality for screen reading. That combination removes most of the excess size while staying visually indistinguishable from the original for the vast majority of documents.
Why is my scanned PDF so much bigger than a similar text document?
Because every page is a full-resolution image rather than lightweight text instructions. A single scanned page can be larger than an entire chapter of typed text.
Does removing metadata actually save much space?
Rarely more than a few kilobytes on its own — it matters for privacy far more than for file size. Images are almost always where the real savings are.
Can I compress a PDF without uploading it anywhere?
Yes — the Compress PDF tool on this site re-encodes images entirely in your browser using the same techniques described here, with the file never leaving your device.