What Is PDF Compression? How PDFs Are Made Smaller

Lossless vs Lossy Compression

Lossless compression reduces file size without any data loss — the decompressed output is byte-for-byte identical to the original. It works by finding patterns and redundancy in data. Used for: text streams, font programs, vector graphics, and images where quality loss is unacceptable. Common algorithms: Flate (ZIP/DEFLATE) and LZW. Lossy compression achieves much higher compression ratios by discarding data that is imperceptible (or acceptable to lose). Used for: raster images (photographs). Common algorithms: JPEG and JPEG 2000. The quality degradation from lossy compression is a trade-off between file size and visual fidelity.

What Gets Compressed in a PDF

• Image streams: the largest contributor to PDF size. Photos and scanned content benefit most from JPEG compression at 70-80% quality — often reducing image data by 80-90% with minimal visible degradation.
• Content streams: text drawing operators, path commands, color settings. Compressed with Flate — typically achieves 50-70% reduction.
• Font programs: embedded font data. Compressed with Flate; also reducible via font subsetting (embedding only used glyphs).
• Object streams: PDF 1.5+ allows multiple objects to be grouped and compressed together with Flate, significantly reducing overhead for PDFs with many objects.

Image Downsampling: The Biggest Win

The most impactful compression technique for most PDFs is image downsampling — reducing the pixel dimensions of images before compression. A photograph scanned at 600 DPI contains four times as many pixels as the same image at 300 DPI, and sixteen times as many as at 150 DPI. For screen reading, 150 DPI is perfectly adequate; for print, 300 DPI is standard. Downsampling a 600 DPI scan to 150 DPI before JPEG compression can reduce an image from 20 MB to under 200 KB with no visible quality loss at normal reading distances.

Other Compression Techniques

• Remove embedded thumbnails: Acrobat embeds small preview images for each page — removing them saves space with no functionality loss
• Subset fonts: embed only the glyphs used in the document rather than the complete font (saves 50-90% on font data)
• Remove duplicate resources: identical images embedded multiple times can be deduplicated
• Linearization: reorganizes for web performance — doesn't reduce size
• Strip metadata: removing XMP metadata and DocInfo saves a small amount

Using FixMyPDF to Compress

The FixMyPDF PDF compressor applies Flate compression to content streams, JPEG compression to images at optimized quality, font subsetting, and duplicate resource removal. It runs entirely in your browser — your file is never uploaded to a server. For most PDFs with embedded images, expect 30-70% file size reduction. For PDFs containing only text and vector graphics, compression gains are more modest (10-30%) since text content is already efficiently stored.