How to Compress a PDF Without Losing Quality (Complete 2026 Guide)

A PDF that opens at 15MB but contains only a few pages of text and simple diagrams is an avoidable problem. PDF files become unnecessarily large for reasons that are almost always fixable — you just need to know which cause applies to your file and which tool addresses it. This guide covers every technique available in 2026 with realistic expectations for each.

What Makes PDFs Large: The 5 Main Causes

Before compressing, it helps to know why your PDF is large. The five main contributors are not equally fixable, and misidentifying the cause leads to disappointed results.

1. Embedded High-Resolution Images

Photos embedded in PDFs are the single largest contributor to file size. A 12-megapixel camera photo saved at 100% JPEG quality is around 8–15MB. An architectural drawing scanned at 600 dpi might be 20MB per page. When a PDF contains even a few high-resolution photos, the file size reflects the aggregate of those images, regardless of how few pages there are.

The key variable is the compression applied to images within the PDF structure. PDFs support JPEG (lossy, ideal for photos), JPEG2000 (better quality-to-size ratio), PNG/ZIP (lossless, best for line art), and CCITT/JBIG2 (monochrome scans). Many applications embed images uncompressed or at unnecessarily high quality settings.

2. Incremental Save History

Every time you edit a PDF and save it without choosing "Save As" or "Export", most PDF editors append the changes to the end of the file rather than rewriting it entirely. This is by design — it allows undo functionality and preserves digital signatures — but after many edits, a document can contain 5–10 old versions of pages that are never displayed but occupy space. "Flattening" the file by re-exporting collapses all history into a single clean version.

3. Embedded Fonts

PDFs embed font data to ensure correct rendering on any device that opens the file. A complete font with all Unicode glyphs can be 500KB to 3MB. If a document uses 5 different typefaces and embeds complete fonts (rather than subsets), that's 2.5–15MB in font data alone. Proper subsetting — embedding only the characters actually used — dramatically reduces this overhead.

4. Metadata and XMP Bloat

Some PDF creation tools write extensive metadata: revision history, creation workflow data, thumbnail images, ICC colour profiles, and extended XMP packets. While useful for professional workflows, this metadata is often unnecessary for documents being shared as final deliverables and can add hundreds of kilobytes.

5. Unoptimised Object Streams

The internal structure of a PDF (its cross-reference tables and object streams) can be stored in a compressed or uncompressed format. PDF versions 1.5 and later support cross-reference stream compression, which can reduce structural overhead by 20–40% in large documents. PDFs saved with older compatibility settings or by poorly-optimised tools miss these savings.

Method 1: Browser Compression — Fast, Private, Free

Folium's Compress PDF tool processes your file in your browser using pdf-lib. It strips unnecessary metadata, flattens incremental update history, and applies object stream compression. All processing is local — your file is never uploaded to any server.

1. Open foliumio.netlify.app/compress-pdf.html.
2. Upload your PDF by dragging it or clicking Choose File.
3. Choose your compression level. Low preserves all metadata and applies only structural compression — ideal for archival documents where metadata matters. Medium strips redundant metadata and optimises object streams — the right choice for most documents. High removes as much metadata as possible while maintaining visual fidelity.
4. Click Compress PDF. The tool shows you the exact before/after file size and percentage saved before you download.

Realistic results by file type:

• Text-only PDFs with metadata bloat: 20–50% reduction
• PDFs with many incremental edits: 30–60% reduction
• Image-heavy PDFs already optimised at creation: 5–15% reduction
• Scanned PDFs (image-only): 5–20% reduction

Method 2: Remove Unnecessary Pages

Sometimes the most effective "compression" is removing content you don't need. Consider:

• Blank pages inserted by scanners (often at the start or end)
• Cover pages not needed in the shared version
• Full-page advertisements in downloaded reports (common in industry publications)
• Large appendices when sharing only the main body of a report

Use Remove Pages to specify exactly which pages to delete. For a 40-page report where pages 35–40 are an appendix you don't need, enter 35-40 and those pages are removed instantly.

Method 3: Adobe Acrobat PDF Optimiser

Acrobat Pro's PDF Optimiser provides the most granular compression control available in any consumer tool. Before applying any compression, click Audit Space Usage to get a breakdown of what's consuming space — this tells you exactly where to focus your optimisation effort.

1. Open the PDF in Acrobat Pro.
2. Go to File → Save As Other → Optimised PDF.
3. Click Audit Space Usage to see the space breakdown.
4. Adjust settings: for image-heavy files, focus on the Images panel — set colour images to JPEG compression at 150 dpi for screen, 300 dpi for print. For text-heavy files, the Discard Objects and Clean Up panels offer the best savings.
5. Click OK and save the optimised version.

Acrobat Optimiser consistently achieves better results than browser-based tools for image-heavy documents because it can re-encode images at lower DPI — something client-side tools cannot do without significantly more processing.

Method 4: Image Recompression Pipeline

If your PDF is large primarily because of embedded photos, the most effective approach is to extract the pages as images, compress those images, and reassemble them into a new PDF. This pipeline can reduce heavily image-laden PDFs by 50–80%.

1. Use PDF to JPG to extract all pages as JPEG images. Pages are rendered at 2× scale for sharp output.
2. Compress the JPG images using a tool like Squoosh (free, browser-based), TinyJPG, or macOS's Preview (Export with reduced quality). Aim for 80–85% JPEG quality for most documents — the difference between 85% and 100% is invisible at reading distances but saves 60–70% of file size.
3. Convert the compressed images back to PDF using JPG to PDF. Select all the compressed images in order and choose A4 or Letter page size.

This pipeline produces an image-only PDF (no text layer). If you need the text to remain searchable, OCR software is required after reassembly.

Special Case: Compressing Scanned PDFs

Scanned PDFs are entirely image-based, making them the hardest to compress without visible quality degradation. A few specific strategies that work well:

• Colour to greyscale conversion: If the original document is black text on white paper, converting the images to greyscale before embedding reduces file size by 60–70% compared to colour. Use any image editor to convert.
• DPI reduction: 150 dpi is sufficient for most text documents viewed on screen. 300 dpi is the professional standard for printing. If your scan was captured at 600 dpi, downsampling to 300 dpi cuts the image data in half with no visible quality change for printed output.
• JBIG2 for black-and-white scans: JBIG2 is a specialised compression format for monochrome images that can achieve 10–20KB per page for clean text scans. Acrobat Pro and some scanner software support this format.

Creating Smaller PDFs at the Source

The best time to optimise PDF size is during creation, not after. Configuration advice by application:

Microsoft Word: File → Export → Create PDF/XPS → Options. Set to "Standard" for general use, or uncheck "Bitmap text when fonts may not be embedded" to avoid unnecessary rasterisation. In Word 365, the export dialog also allows image quality settings.

Adobe InDesign: Use the "Smallest File Size" PDF preset for digital distribution, which applies JPEG compression to images and disables printer marks. For print-ready PDFs, use PDF/X-1a.

Google Docs: Google's PDF export is reasonably optimised for text documents. For image-heavy documents, the output can be large — compress after export using Folium or Acrobat.

macOS Print to PDF: macOS's built-in Print to PDF is efficient for text documents. The Quartz filter "Reduce File Size" in ColorSync Utility can be applied for additional compression, though it aggressively downsamples images.

Common Compression Mistakes to Avoid

Using Print to PDF as a compression method: This often re-renders the document at screen resolution, converting crisp vector graphics and text to raster images. The result may actually be larger, or the same size but with reduced quality.

Multiple compression passes: Running a PDF through compression multiple times rarely produces additional savings after the first pass. Repeated JPEG re-compression degrades image quality with each pass.

Uploading sensitive documents to cloud compression services: Many free online PDF compressors are server-based. Your confidential contracts, financial statements, or personal documents may remain on their servers indefinitely. Use local tools for any sensitive files.

Compression Method Comparison

Method	Best For	Typical Reduction	Preserves Text	Privacy
Folium	Metadata/structure bloat	15–50%	Yes	Local
Remove Pages	Removing content	Variable	Yes	Local
Acrobat Optimiser	Image-heavy, granular control	20–70%	Yes	Local
Image pipeline	Photo-heavy PDFs	50–80%	No (images only)	Local