How To Compress PDF Files Without Losing Usability

This guide maps to the tool directly so you can apply each step while reading.

Understand what makes a PDF large before compressing #

PDF file size is driven by three main things: embedded images, embedded fonts, and metadata. Of these, images are responsible for the largest sizes by far. A scanned contract or a brochure exported from a design tool can easily reach 10 to 20 MB because each page was saved as a high-resolution image at print quality — far higher than what a screen or standard printer needs.

Text-heavy PDFs generated from word processors are typically much smaller because they store text as actual characters rather than images. If your PDF was generated from LibreOffice, Word, or a web-based converter, it may already be close to optimal. Compressing it further may yield minimal savings and, in some cases, can marginally degrade embedded font rendering.

Scanned documents occupy a category of their own. Each page is literally a photograph of paper. Compression on these files can produce substantial savings — sometimes reducing a 15 MB scan to under 2 MB — without making the text unreadable, as long as the original scan resolution was reasonable. 150 to 300 DPI is the common range for archival and practical use.

Knowing what type of PDF you have before compressing helps you set realistic expectations. An image-heavy brochure and a text-heavy report will not behave the same way under compression, and the right approach differs between them.

Prepare files correctly before compressing #

Always compress the final version of a document, not a draft. If you know the document will be edited again, wait. Re-compressing an already-compressed PDF can introduce compounding quality loss, especially in image-heavy files, and makes troubleshooting harder when output does not look right.

If you have multiple documents that belong together in a submission, decide whether to merge them first or compress each separately. For portal submissions requiring one single file, compress each source file individually first, then merge the compressed results. Merging uncompressed files first and then compressing the combined document works but gives you less control over how each section looks.

For scanned documents, check the scan quality before compressing. A scan taken at very low DPI — below 100 — will look poor before and after compression. In that case, re-scanning at 150 to 200 DPI and then compressing will produce better results than trying to salvage a low-quality input through heavy compression.

Rename files clearly before you start. Working with names like report-compressed.pdf and report-original.pdf keeps the process clean, makes it easy to compare versions, and ensures you never accidentally send the wrong version to a recipient.

Validate compressed output carefully #

After compressing, the first check is always legibility. Open the compressed PDF and zoom to 100 percent on a text-heavy page. If body text looks soft or blurry, the compression was too aggressive for this type of document. For scanned text, acceptable compression means you can still read every word at normal viewing size without squinting.

Check on a mobile device too. Many documents will be opened on phones, and mobile screens amplify compression artifacts. A file that looks fine on a desktop at 100 percent zoom can look noticeably soft on a phone's high-DPI screen. A quick thirty-second check on your phone before sending is worth the effort.

For documents with tables, check that borders, cell lines, and small-text headers inside the table are still clear. These fine details are the first to degrade under heavy compression and are the most noticeable to recipients who rely on them for navigation.

If you are submitting to a portal with strict size requirements, verify the file size matches expectations and test the upload flow before the submission deadline. Some portals also have maximum page dimension limits alongside file size limits, which compression alone will not fix.

When compression alone is not enough #

If a compressed PDF is still too large, the most effective next step is usually to address the source material rather than compress harder. Look at which pages are contributing most to file size. If a few pages contain very large embedded screenshots or photos, replacing those images with lower-resolution versions in the source document and regenerating will produce better results than repeating compression.

Cropping unnecessary whitespace from images before generating or scanning the document also helps significantly. A full-page scan of an A4 sheet to capture a signature block that occupies one-eighth of the page is wasteful — cropping to the relevant portion before scanning reduces size more effectively than post-compression.

If the PDF contains pages you do not actually need to share, use the split tool to extract only the relevant pages and compress that smaller subset. Sending a 3-page extract compressed is more reliable than trying to compress a 40-page document into an email-friendly size.

For cases where a scanned document simply cannot compress small enough without becoming unreadable, consider creating it as a properly prepared image-to-PDF from the start, using images already optimized for file size before conversion. This gives you direct control over quality and size from the beginning.