FileNorm
← Back to the blog
2026-07-20

How to Reduce PDF File Size Without Losing Quality

The real ways to shrink a PDF — image downsampling, font/font-subset handling, and re-compression — explained plainly, with the tradeoffs each one actually makes.

Most PDFs aren't heavy because of text — text is tiny. They're heavy because of the images embedded inside them: a scanned document, a photo pasted into a page, or a diploma photographed at full camera resolution and dropped straight into a PDF. Reducing "PDF size" in practice almost always means reducing the size of the images inside it.

What actually makes a PDF small

There are three real levers:

  1. Re-compress the embedded images. A photo saved at 90%+ JPEG quality inside a PDF is often 3–10x larger than it needs to be for something that just needs to be legible on screen or in print. Re-encoding those images at a lower (still visually fine) quality is where most of the savings come from.
  2. Downscale oversized images. A photo from a modern phone camera is routinely 3000–4000px wide. A PDF page rarely needs more than 1500–2000px of resolution for anything printed or read on screen. Cutting the pixel dimensions in half cuts the data to roughly a quarter.
  3. Strip what you don't need. Unused embedded fonts, leftover metadata, and duplicate image data (common when a PDF has been edited or merged multiple times) all add dead weight that compresses away with zero visible difference.

What doesn't actually help

Re-saving the same PDF from a different app rarely does much — most tools use similar default settings, so a Word-exported PDF re-saved from Acrobat looks about the same size. "Print to PDF" from a browser can occasionally make things worse, because it may re-rasterize already-compressed content at a higher bit depth than the original.

The practical way to do it

If the PDF needs to hit a specific size — a portal that rejects anything over 1MB, an email attachment limit, a form with a stated cap — trial-and-error with quality sliders wastes time. A target-size compressor does the same re-compression and downscaling work automatically, searching for the highest quality that still lands under the number you need:

All of this runs in your browser — the PDF is never uploaded anywhere, which matters if it contains anything personal (IDs, diplomas, signed forms).

A quick sanity check before you submit

After compressing, open the result and actually look at page 1 at 100% zoom, not just the thumbnail. Text should stay sharp (text isn't touched by image compression) and photos should still be clearly legible — some visible smoothing on a scanned signature or photo is normal and expected at aggressive size targets, but blur that makes text in a scanned document hard to read means you compressed too hard for that particular file.