Why Won't My PDF Compress?
Updated 8 September 2026
You run a 14MB PDF through a compressor, wait, and get back 13.6MB. Nothing is broken — the tool has done what it can, and the reason it could not do more is sitting inside the file. Almost every case of a stubborn PDF comes down to one distinction, and once you know which side of it your document falls on, the fix is straightforward.
First, work out which kind of PDF you have
Open the file in any PDF reader and try to select a line of text with your cursor. If individual words highlight, you have a text-based PDF — it stores actual characters, and its size is mostly structure that optimisation can strip. If the whole page selects as one block, or nothing selects at all, you have a scan: every page is a photograph of paper. This thirty-second test tells you more than any file size does, and it determines everything that follows.
Why scans resist compression
A scanned page is an image, and your scanner already compressed it on the way out — usually as JPEG inside the PDF wrapper. A general PDF optimiser can tidy the document's internal structure, remove duplicated fonts, and clean up metadata, but it cannot re-encode those page images without decoding and re-compressing every one of them. Browser-based tools in particular cannot do this at all. So you are left optimising the few percent of the file that is not image data, which is why the number barely moves.
The fix that actually works: rescan
If you still have the original document, rescan it with two settings changed. Drop the resolution from 600 DPI to 200 DPI, and switch from colour to greyscale or black-and-white if the page is dark text on white paper. Each of those roughly quarters the data independently, so together they routinely produce a file five to ten times smaller — and at 200 DPI, printed text remains completely legible. Most people scan at high colour resolution by default without ever choosing to, and end up storing the paper's texture and yellowing in fine detail.
If you cannot rescan
When the scan is all you have, your options are narrower. Removing pages you do not need helps proportionally — extracting the three relevant pages from a forty-page bundle cuts the size by more than any compressor will. Converting the PDF's pages to JPG, compressing those images, and rebuilding a PDF from them works but is fiddly and loses any text layer. The genuinely effective route is server-side compression built to re-encode embedded images, which is a different class of tool from anything that runs in a browser tab.
Check the limit before you optimise for it
A surprising number of compression problems dissolve on inspection. People assume a portal wants 100KB when it accepts 2MB, or compress a file that was already under the limit. Read the actual requirement on the upload page first. And when you do compress, open the result and read the smallest text on the page before submitting — a file that technically uploads but cannot be read by whoever reviews it has failed in a more expensive way than one that was slightly too large.
Frequently asked questions
- My PDF got bigger after compressing. How?
- Some PDFs are already stored optimally, and re-saving adds structural overhead without removing anything. If the output is larger, keep your original — there was nothing to gain.
- Does compressing a PDF reduce its text quality?
- No. Text in a PDF is stored as characters and vectors, so it prints at full quality regardless of file size. Only embedded images can lose quality.
- How do I know if my scan is at 600 DPI?
- Check your scanner software's settings rather than the file. As a rough guide, a single A4 page scanned at 600 DPI in colour is often 5MB or more, while 200 DPI greyscale is typically under 500KB.