How to Compress a PDF File Size: What Actually Works and Why
PDF files can balloon in size faster than you'd expect. A single scanned document might hit 50MB. A presentation exported from PowerPoint can easily push past 20MB. Understanding why PDFs get large — and what compression actually does — helps you pick the right approach for your situation.
Why PDF Files Get So Large
PDFs are container files. They can hold text, fonts, vector graphics, raster images, embedded metadata, digital signatures, form fields, and more. Each element adds weight.
The biggest culprits are almost always images. When you scan a document, each page becomes a high-resolution image embedded inside the PDF. A 300 DPI scan of a 10-page document can easily produce a 15–30MB file. Fonts embedded for display consistency add overhead. Layered graphics from design tools add more.
Text-only PDFs — a simple contract or a few pages of typed content — are almost always small by nature. The problem almost exclusively lives with image-heavy files.
What PDF Compression Actually Does
"Compressing a PDF" isn't one thing. It's a label that covers several different technical operations:
- Image resampling (downsampling): Reduces the resolution of embedded images — for example, from 300 DPI to 150 DPI or 72 DPI. This is the single biggest lever for reducing file size and involves real quality tradeoff.
- Image recompression: Re-encodes images using more efficient compression algorithms (like JPEG or JPEG 2000 for photos, or CCITT Group 4 for scanned black-and-white documents). Quality and size both depend on compression level chosen.
- Font subsetting: Removes unused characters from embedded fonts. If a document uses only 40 characters of a 500-character font, only those 40 need to be stored.
- Removing hidden data: Strips metadata, comments, hidden layers, embedded thumbnails, or form submission histories that add size without visual content.
- Flattening: Merges transparency layers and simplifies complex graphics objects.
Most compression tools apply some combination of these — and the proportions vary significantly between tools and settings.
The Main Ways to Compress a PDF 🗜️
Online Compression Tools
Browser-based tools (like Smallpdf, ilovepdf, Adobe's online compressor, and similar services) handle compression server-side. You upload your file, the tool processes it, and you download the result. These are fast and require no software installation.
What to consider:
- File privacy — your document is uploaded to an external server
- File size limits — free tiers often cap uploads at 5–25MB
- Compression control — many offer only basic presets (low/medium/high) without granular control
- Output consistency — results vary noticeably between tools even at the same quality setting
Desktop Software
Applications like Adobe Acrobat Pro, Preview on macOS, PDF24, and LibreOffice Draw compress PDFs locally on your device. No upload required.
Adobe Acrobat Pro's PDF Optimizer gives the most granular control — you can set image resolution targets per image type (color, grayscale, monochrome), choose compression codecs, and selectively strip specific data categories. It's powerful but requires a subscription.
macOS Preview offers a quick "Reduce File Size" export option using a Quartz filter. It works, but applies aggressive and somewhat unpredictable compression — effective for casual use, less so when output quality matters.
PDF24 is a free desktop app for Windows that offers a middle ground: more control than Preview, no subscription required.
Compression at the Source
The most underrated approach: reduce file size before the PDF is created.
- Export images at lower resolution before inserting them into your document
- Use "Optimize for web" or "Compress pictures" settings inside Word, PowerPoint, or Google Slides before exporting
- Scan documents at 150–200 DPI instead of 300+ DPI when high resolution isn't needed
Source-level reduction almost always produces better quality-to-size ratios than post-compression, because you're making intentional choices rather than letting an algorithm guess.
How Much Can You Actually Reduce File Size?
It depends entirely on what's in the file.
| File Type | Typical Compression Potential |
|---|---|
| Scanned image-heavy PDF | Often 50–80% reduction |
| Photo-rich presentation PDF | Often 40–70% reduction |
| Mixed text and image PDF | Often 20–50% reduction |
| Text-only or vector PDF | Often 5–15% reduction |
These are general patterns, not guarantees. A scanned PDF with already-compressed JPEG images may shrink very little on a second pass. A PDF with uncompressed TIFF images embedded may compress dramatically.
Quality vs. Size: The Tradeoff That Matters Most 📄
Every compression decision sits on a spectrum between file size and visual fidelity. There's no free lunch.
For internal documents you're sharing quickly by email, aggressive compression — dropping image resolution to 96 DPI — is usually fine. Nobody's zooming in.
For client-facing materials, print-ready files, or documents with fine text detail in images (like scanned technical drawings or legal documents), aggressive compression can make text blurry or images artifact-heavy. You need to test the output.
Where your situation matters:
- Are you compressing for email attachment limits, or for long-term archival?
- Does the PDF contain scanned text that needs to remain legible?
- Is privacy a concern that rules out online tools?
- Do you need batch compression across hundreds of files, or just one?
- Are you on Windows, macOS, iOS, or Linux — and which tools are available to you?
The right compression method and the acceptable quality level depend on answers that are specific to your workflow, your audience, and what the document is actually used for. That's the piece no general guide can settle for you.