How to Compress a PDF to 200KB: What Actually Shrinks and What Doesn't

Compress a PDF under 200KB in your browser without uploading it. Learn which types of PDFs shrink well, which don't, and how compression tools actually work behind the scenes.

7 min read

The 200KB target and why it keeps showing up

A surprising number of upload forms cap PDF attachments at 200 KB. Government tax portals, university application systems, some bank KYC uploads, and old-school corporate document management platforms all seem to have converged on this number. It made sense in 2005; it makes less sense today. But if a form rejects your file, no amount of arguing with the form fixes the file.

This guide walks through what actually gets smaller when you compress a PDF, when 200KB is realistic, and when you need a different strategy entirely.

What a PDF is made of, size-wise

A PDF is a container that holds text, fonts, images, vector graphics, and metadata. In most everyday documents, roughly:

  • Embedded raster images: 60-90% of the file size for scans and photo-heavy documents; 20-40% for typical business documents with a logo or two
  • Embedded fonts: 5-30% depending on how many fonts and whether they were fully embedded or subsetted
  • Text and vector graphics: usually under 5%
  • Metadata, structure, cross-reference tables: 1-3%

The takeaway: if your PDF is big, it is almost always because of images. Compressing a text-only 400 KB PDF down to 200 KB is easy. Compressing a 5 MB scan-heavy PDF to 200 KB is a different problem that may require re-scanning at a lower resolution or accepting readable-but-fuzzy output.

How browser-based compression actually works

Our compression tool runs pdf-lib in your browser to rewrite the PDF with lower-cost representations of its contents. Three things happen in sequence:

Image re-encoding. Every raster image in the PDF is decoded, then re-encoded as JPEG at a lower quality setting (75, 60, or 40 depending on the compression level you picked). A 2 MB PNG logo can drop to 80 KB with barely visible quality loss because most logos have flat colours that JPEG handles efficiently.

Font subsetting. If a font is embedded but only a few characters from it are used (common when a PDF was generated by rendering a webpage that references twenty fonts, half of which appear only in alt text), the tool strips the unused glyphs. A full CJK font can be 6-10 MB; a subset covering just the actually-used characters is usually 40-200 KB.

Object deduplication. PDFs occasionally contain the same image or resource embedded multiple times because the generator did not deduplicate. Compression tools detect identical byte streams and replace duplicates with references, so the resource is stored once.

None of these operations involve the network. Your file stays in your browser tab.

Realistic compression ratios by document type

Based on running our compression tool on a variety of real documents, here is what to expect at the medium compression level:

Content type Typical ratio Notes
Text-only reports, articles 50-70% smaller Fonts subset, whitespace normalised
PowerPoint exports with logos and charts 40-60% smaller Logo images JPEG-encoded
Colour scans at 300 DPI 60-80% smaller Biggest gains, quality drops modestly
Colour scans at 600 DPI 70-85% smaller Even bigger gains, 600 DPI is overkill for most uses
Photo-heavy PDFs (portfolios) 30-50% smaller Photos already JPEG-compressed on input
Already-optimised PDFs 0-15% smaller Some files are already close to optimal
Vector-only technical drawings 5-20% smaller Little to compress; no bitmap data

The "already-optimised" row is where people get frustrated. A PDF that was generated by a good tool with reasonable JPEG quality already may not shrink much further, no matter what compression level you pick. If you see a compression ratio under 20%, chances are the file is already close to its optimal size.

Getting to 200KB specifically

Some strategies stacked in order of effort:

Try the "high" compression level first. For most PDFs this alone gets you 40-70% reduction. Text stays readable, images look slightly softer.

Reduce the source before generating the PDF. If you scanned a document at 600 DPI, rescan at 300 DPI. If you exported a Word doc with 4K images, downsize them in the source doc first. This gives much better results than any post-hoc compression, because you can control the trade-off exactly.

Split into fewer pages. A 10-page PDF that needs to hit 200 KB gives you 20 KB per page, hard. A 3-page PDF at 200 KB gives you 66 KB per page, comfortable for text, tight for scans.

Convert to greyscale. Colour scans re-encoded as greyscale drop 40-60% more. Most tools do not have this option built in; you can work around it by opening the PDF in an image editor, converting each page to greyscale, and re-exporting.

Split the document and send in parts. Rarely elegant, but for truly image-heavy documents that must stay legible, splitting into 2-3 smaller PDFs each under 200 KB is more practical than trying to squeeze everything into one.

What browsers do and don't handle well

Because compression happens entirely in your browser, browser memory limits matter more than usual:

  • Desktop Chrome, Edge, Firefox: comfortable with PDFs up to ~200 MB source size
  • Safari on macOS: reliable up to about 120 MB before running out of memory
  • iOS Safari and iPad Safari: 40-60 MB is the practical ceiling because iOS caps per-tab memory aggressively
  • Android Chrome: varies wildly by device, but 50-80 MB is a safe target

If your source PDF is above these limits, the compression tab either crashes or runs out of memory partway through. Workarounds: compress on desktop and email the result to your phone, or split the PDF into smaller pieces and compress each piece separately.

Things that will not fit into 200KB

Being honest about limits:

  • A 50-page colour magazine at print quality: no amount of browser-based compression will fit this into 200 KB while keeping it legible.
  • Scans of hand-drawn technical diagrams that need high resolution to remain readable: JPEG artifacts destroy fine detail.
  • Multi-page contracts with embedded photos: usually you can hit 500-800 KB but not 200 KB without visibly degrading the photos.

If a form requires 200 KB and your file can't get there without becoming unreadable, the pragmatic move is to ask the form owner whether they accept a larger file, or split the document and upload in parts if the platform allows it.

Frequently asked questions

How much can I actually compress a PDF?

Between 0% and 90%, depending on the source. Photo-heavy documents at high resolution compress the most (60-85% is typical). Text-only PDFs already generated efficiently might only shrink 10-20%. There is no universal "compress by X%", the tool works with what the source gives it.

Form fields, links, and bookmarks are preserved by our compression tool. Digital signatures (PKI-based) will be invalidated because compression rewrites the file bytes and any cryptographic hash of the file changes. Visual signature images are just embedded raster images and are re-encoded at the chosen JPEG quality, they may look slightly softer after compression but are still recognisable.

Why did my file not shrink at all?

The PDF was probably already optimised. This is common for files that came out of professional design software like InDesign or Illustrator, from PDF/A archival tools, or from previous rounds of compression. There is a compression floor for every file, and if you are already at it, the tool honestly cannot do more.

Can I compress a password-protected PDF?

Only after removing the password. Compression tools need to read the page contents to rewrite them, and encrypted pages are unreadable until the password is applied. Decrypt first, then compress.

Does aggressive compression damage text?

Text in a PDF is stored as characters and font references, not as an image. Compression does not affect text quality regardless of the compression level. It affects images only. If your PDF's text looks fuzzy after compression, the text was already an image in the source (which happens with scanned documents or exports from some Word-to-PDF tools).

Is my file uploaded when I compress it in the browser?

No. Our tool loads the PDF into your browser's memory, rewrites it using pdf-lib, and offers the result as a download. Closing the tab discards everything. If your network shows uploads while you are compressing, that is not the compression tool, probably ad scripts or analytics unrelated to the PDF itself.