A bloated PDF does not have to mean an ugly PDF. This guide walks you through genuinely lossless shrinking methods first, then shows how to downsample images intelligently — so your file gets dramatically smaller while looking exactly the same at normal size.
Not all compression is equal. There are two fundamentally different kinds, and knowing the difference is the key to shrinking a PDF without visibly degrading it. Lossless compression removes wasted bytes without touching a single visible pixel or character: deleting pages you do not need, stripping hidden metadata, subsetting fonts, and dropping annotations. Do these first, because they cost you literally nothing in quality. Lossy compression, on the other hand, throws away visual information — usually by downsampling images or lowering JPEG quality. Done carefully, lossy compression can shrink a file by 70–90% while remaining perfectly readable; done carelessly, it turns crisp text into mush.
The honest rule of thumb: if your document is mostly text (contracts, reports, ebooks), you can usually cut 10–40% with purely lossless methods. If it is mostly images (scans, photo portfolios, catalogs), lossless methods help less, and you will need smart, controlled image downsampling — which this guide shows you how to do without the result looking bad.
Work through these in order. Each one removes bytes that contribute nothing to what you see on screen or on paper.
For image-heavy PDFs, this is where the real savings live. The trick is matching resolution to how the document will actually be viewed — anything beyond that is wasted megabytes:
Only needed if the PDF goes to a commercial printer. For a file you will email or read on screen, 300 DPI images are pure bloat.
The sweet spot for most documents. Text stays crisp, photos look great, and files are a fraction of their 300-DPI size.
Perfect for PDFs that live on phones and laptops. A scanned page at 100 DPI is still perfectly readable and can be 80% smaller than the 300-DPI original.
Downsampling only helps image-based pages. If your PDF already has vector text, re-rendering it as images (some “compressors” do this) can make quality worse and the file bigger.
Our free in-browser compress tool lets you pick the target quality yourself and shows honest before/after sizes — your file never leaves your device, so it is safe for confidential documents too.
JPEG artifacts stack. Each recompression round throws away more detail. Always compress from the highest-quality original you have.
Virtual printers rasterize text into images — the opposite of what you want. Use the app’s native Export/Save-as-PDF instead.
The difference between 80% and 95% compression is often 3× the file size for barely visible gain. Moderate settings win.
Fine print, charts, and small diagrams suffer first. A 30-second zoom check catches problems before you send the file to a client.
Numbers make the abstract concrete. Here is what each approach typically achieves on a 10-page document, so you can sanity-check your own results:
Before: 8.2 MB (full fonts embedded, print-to-PDF). After lossless pass: 1.1 MB — subset fonts, stripped metadata, native export. Visible difference: none whatsoever.
Before: 46 MB at 600 DPI. After 150-DPI downsample: 7.4 MB. Visible difference: none at normal reading size; tiny softness only past 300% zoom.
Before: 88 MB (300-DPI originals). After 150-DPI + 80% JPEG: 14 MB. Visible difference: imperceptible on screen; acceptable for office prints, not for gallery prints.
If your “compressed” file is bigger than the original, stop: the tool likely re-rendered vector text as images. Undo, and use a tool that preserves text as text.
The pattern: lossless methods first (always safe), then one careful image pass matched to the document’s real use. When the before/after looks identical at 100% zoom, you have compressed without losing quality — by definition.
Yes — with lossless methods. Removing unneeded pages, stripping metadata, flattening forms, subsetting fonts, and dropping annotations all shrink the file without changing a single visible pixel. For text-heavy documents these alone often save 10–40%.
Match it to the use: 300 DPI for professional printing, 150 DPI for office printing and screen reading, 72–100 DPI for screen-only documents. Most emailed PDFs are perfect at 100–150 DPI.
Usually one of three causes: the images were downsampled too aggressively, the file was compressed more than once (JPEG artifacts stack), or a tool re-rendered vector text as images. Recompress once, from the original, at a gentler setting.
Uploading sensitive documents to a random website is risky. Prefer tools that process locally — every tool on this site runs 100% in your browser, so your file never leaves your device.
It depends on the content. Text-heavy PDFs: 10–40% via lossless methods. Scanned or photo-heavy PDFs: 60–85% with careful 100–150 DPI downsampling that remains perfectly readable on screen.
No. Compression and encryption are independent. If you need to remove a password you know, use a dedicated unlock tool; compression alone will not decrypt anything.
Compress after merging if the pages share similar content — one consistent pass gives uniform quality. If the source files differ wildly, compress each appropriately first, then merge.