Strips metadata and enables cross-reference streams for smaller file sizes.
How it works
The PDF Compress Preset tool applies structural compression to a PDF document by enabling cross-reference streams (the pdf-lib useObjectStreams flag) and stripping embedded metadata fields that occupy space without affecting document content. The result is a smaller PDF that renders identically to the original.
What metadata is removed: - Title, Author, Subject, Keywords — document property fields visible in File → Properties - Creator (the application that created the PDF) - Producer (the PDF library that generated it)
These fields are often set to lengthy strings by the authoring application and serve no functional purpose in distributed documents. Removing them typically saves a few hundred bytes to a few kilobytes, though savings vary.
Cross-reference stream compression (the primary size reduction mechanism): PDF cross-reference tables index the byte offset of every object in the file. Traditional PDF files use ASCII xref tables. Cross-reference streams use binary encoding and compression, typically reducing the index overhead by 30–60%.
When this tool is most effective: - PDFs generated by non-optimizing tools (some older LaTeX distributions, custom PDF generators) - PDFs that have been repeatedly edited, accumulating orphaned object references - PDFs where the cross-reference table was never compressed
When this tool provides minimal benefit: - PDFs already saved with optimization (Acrobat "Save as Optimized PDF", Ghostscript with PDFSETTINGS=/ebook) - PDFs with very large embedded images (image data dominates — use a specialized image compressor)
For maximum size reduction on image-heavy PDFs, consider also running the PDF through an image optimization pass.
Privacy: all processing uses pdf-lib in the browser. File size data is displayed locally.
Frequently Asked Questions
- It depends on how the PDF was generated. PDFs created by older tools, custom generators, or applications that don't optimize their output can see 10–40% size reduction. PDFs already saved with Acrobat's 'Optimize' or Ghostscript's PDFSETTINGS=/ebook flag see little to no change — the tool reports 'no reduction' in this case. The biggest gains come from enabling cross-reference streams on older PDF 1.3/1.4 files that pre-date this compression method.
- No. This tool does not recompress or downsample images. It only affects the PDF's structural data (cross-reference tables and metadata) and document property strings. Image quality is completely preserved. For image-driven size reduction, you would need an image optimization step before creating the PDF, or a tool like Ghostscript with downsampling options.
- Title, Author, Subject, Keywords, Creator, and Producer fields are cleared. This metadata is visible in File → Properties in any PDF reader. It is not recoverable after clearing — the tool discards it. If your organization requires specific metadata on documents (for document management system indexing, for example), do not use this tool, or re-add the metadata after compressing using the PDF Metadata Editor tool.
- File size comparisons depend on whether your file explorer reports the raw file size or the 'size on disk' (which rounds to the nearest disk cluster, usually 4KB). Compare raw byte sizes for an accurate measurement. The tool displays the exact before/after byte count so you can see the true change.