Reduce PDF size: preserve text or compress image pages?

Why can PDF files become so large?

A PDF can contain fonts, vector drawings, selectable text, scanned images and embedded resources. Photo-heavy scans may be much larger than text-first files, while already-optimized documents may have little redundant data left to remove.

Check the number of pages, original file size and whether text can be searched or selected. This helps decide if compression should preserve the document structure or make a visual-only copy.

What is the difference between the two modes?

PangPang Tools offers a preserve-text structural save. It rebuilds the PDF using object streams but does not rasterize page content. Reductions are possible, not guaranteed.

The lossy image-pages option renders every page, exports each as JPEG and creates a new PDF made of those page images. This may reduce the size of image-heavy files, but selectable text, annotations, interactive forms and existing digital signatures can be lost. Keep the original file.

A sensible workflow for reducing file size

Start with the text-preserving option. If the output is smaller and still satisfies the destination’s requirements, inspect and submit it. If the file is mostly scans and structural optimization makes no difference, consider a separate rasterized copy.

The current tool accepts a PDF up to 25MiB. The preserve-text option handles up to 50 pages; image mode handles up to 20 pages. Image mode provides JPEG quality from 10% to 100% and render scales of 0.6, 0.85 or 1. The tool retains the original bytes if the candidate output is not smaller.

• Test the preserve-text mode first
• Keep the source before trying raster mode
• Inspect fine print, diagrams, forms and signatures in the saved copy

Worked example: from 5MB to 3MB

If an original PDF is 5MB and its output is 3MB, the reduction is 2MB or 40%: (5−3)÷5×100. This is an illustrative arithmetic example, not a promised compression ratio.

A smaller file is not automatically an acceptable document. Zoom into labels and table entries, confirm the page order and check whether the recipient needs selectable text or original signed records.

What if the PDF barely gets smaller?

Some PDFs are already compressed. In that case, removing unnecessary pages or separate appendices can work better than another save. For scans, resolution, image complexity and chosen JPEG quality may dominate the output size.

Optical character recognition (OCR) is a separate step that converts text in scanned images into searchable text. This raster compression feature does not perform OCR. Avoid image-only copies when searching, accessibility or signature validity is required.