How to Reduce a PDF from 70MB to 15MB

LazyPDF routes PDF compression through a server running Ghostscript, the industry-standard open-source PDF processing engine. Ghostscript's compression pipeline decodes each embedded image, downsamples it to the target DPI (72 DPI for screen, 150 DPI for balanced quality), re-encodes it as optimized JPEG, and rebuilds the PDF with the smaller image data. Non-image content — text, vector graphics, interactive elements — passes through unchanged. For a 70 MB image-heavy PDF, the screen preset achieves 75–85% size reduction, producing a file in the 10–18 MB range. The balanced (ebook) preset produces 65–75% reduction for a 17–25 MB result with better image quality preservation.

1. 1Go to lazy-pdf.com/compress and upload your 70MB PDF
2. 2The file uploads to the server (expect 20–40 seconds on typical broadband)
3. 3Ghostscript processes the images — this takes 40–80 seconds for a 70MB file
4. 4Download the compressed result — check the output size and open a few pages to verify quality

Understanding how image downsampling achieves compression helps you predict results and make better decisions about compression settings. A photograph embedded in a PDF at 300 DPI for print is approximately 2,480 × 3,508 pixels on an A4 page — about 8.7 million pixels. The same image at 72 DPI is 595 × 842 pixels — about 500,000 pixels. That is a 17× reduction in pixel count, which translates (after JPEG encoding) to approximately a 10–15× reduction in file size. JPEG compression adds another layer of size reduction. A JPEG at quality 85 is typically 60–80% smaller than the same image at quality 100. Combined with downsampling, these two operations together achieve the dramatic file size reductions seen when compressing print-quality PDFs for screen distribution.

1. 1Print-resolution images (300 DPI) → screen resolution (72 DPI): approximately 15-17× pixel reduction
2. 2JPEG quality reduction (100 → 80): approximately 3–5× additional size reduction
3. 3Combined effect: 50–85× total data reduction per image, varying by content complexity
4. 4Net PDF file size reduction: 70–85% for image-dominated PDFs

Some 70 MB PDFs resist standard compression. Possible reasons: the PDF was already compressed — someone generated it with medium-quality JPEG images rather than high-quality originals, leaving little room for further reduction. The PDF contains embedded fonts or binary attachments in addition to images. The PDF uses uncompressible image types for specific reasons (lossless medical images, for example, where JPEG compression is medically inappropriate). For already-compressed PDFs, the options are limited. Trying multiple tools may yield minor variations in output size but not dramatic improvement. The theoretical floor for a JPEG-compressed image PDF is not much below the current compressed state. If you genuinely need the file under 15 MB and standard compression achieves only 35 MB, splitting the PDF in half and sharing in two parts is the most practical solution.

1. 1Try compressing first — check if the output is significantly larger than expected
2. 2Compare compressed sizes from two different tools to verify you are not hitting a tool limitation
3. 3If both tools produce similar output, the PDF content is limiting further reduction
4. 4Split the PDF (lazy-pdf.com/split) into two halves and compress each separately

After compressing from 70 MB to approximately 15 MB, a quick quality check prevents surprises for your recipients. Open the compressed PDF in a viewer and scroll through every fifth page or so, paying attention to: image sharpness at normal reading zoom; text legibility in any text that appears within images (headers, captions, data labels in charts); color accuracy for brand-critical documents where color matching matters; and overall page appearance for anything unexpected. If you find specific pages where quality is unacceptable — typically pages with dense text on an image background or very fine graphical details — note the page numbers. You can extract those specific pages from the original high-quality PDF, compress the remaining pages, and merge the result. This selective compression approach is more work but allows per-page quality control.

1. 1Scroll through the compressed PDF, checking every 5th page at minimum
2. 2Zoom to 150% on image-heavy pages to assess resolution adequacy
3. 3Check that any text overlaid on images is legible at normal reading zoom
4. 4For brand documents, compare colors against the original — compression can shift saturation