How to Compress a PDF to 100KB (or Any Target Size)
'Attachment must not exceed 100KB.' Anyone who has applied for a job, a visa, or a government service online has met a message like this — usually at the last step, with a PDF that is 2MB. The good news: most PDFs carry far more data than they need. The realistic news: whether 100KB is reachable depends almost entirely on what is inside your file.
First, understand where the bytes live. In a typical PDF, text and vector graphics are tiny — a 50-page text-only contract can weigh under 300KB. The heavy freight is images: a single photo page scanned at 300 DPI can be 2-5MB on its own. So the compression strategy is always the same: leave the text alone, and shrink the images.
That is exactly how PdfWill's compress tool works. The 'Recommended' level decodes every embedded image, caps its resolution at what screens and printers actually need, and re-encodes it as an efficient JPEG — while text, fonts, and vector art pass through untouched and stay sharp and selectable. The 'Strong' level pushes the resolution cap and JPEG quality lower for the smallest possible file. Because the work happens in your browser, a 50MB scan doesn't need to crawl through an upload either.
For aggressive targets like 100KB, combine tactics. Strip pages you don't need with the remove-pages tool — the fastest kilobytes you will ever save. If color isn't required, convert the document to grayscale first; color data is expensive. Then run Strong compression. A 10-page text document will land well under 100KB; a 3-page scan will usually make it; a 20-page photo brochure will not, and no honest tool can change that — the pixels simply have to live somewhere.
If you hit the wall, change the shape of the problem instead: split the PDF into parts that each fit under the limit, or check whether the portal accepts a JPG (converting one page with PDF to JPG at moderate quality is sometimes smaller than any full PDF).
One last honesty check that most compressor sites skip: if a tool's output would be larger than your original — which happens with already-optimized files — PdfWill returns the original untouched. A 'compressed' file that grew is a bug, not a service.