Compress PDF to 200 KB
Built for scans. If your PDF is text rather than scanned pages, there is often a better fix than compression.
Updated · Published by Novpix LLC
Drop a PDF here
or click to browse — target is 200 KB
Files are processed on your device and never uploaded.
Runs entirely in your browser — nothing is uploaded
How it works
- 1Work out what you have first: if you can select text in your PDF with the cursor, it is a text document, not a scan.
- 2Drop the file in and click Compress. Pages are re-rendered at a measured resolution until the file lands under 200 KB.
- 3Download the result, or read the reported floor if the document was too long to reach the target.
Frequently asked questions
- How do I tell whether my PDF is a scan or a text document?
- Open it and try to select a word with the cursor. If the text highlights, the PDF stores real characters and is probably already small — a bloated one usually contains an embedded photograph or an oversized logo. If nothing highlights, each page is a picture, which is where compression like this actually helps.
- My PDF is text and still too large. What is making it big?
- Almost always an image inside it: a photograph pasted at full camera resolution, a scanned signature at 600 dpi, or a logo saved as a huge PNG repeated on every page. Compressing those images before they go into the document, or exporting again at a lower image quality from the program that made it, shrinks the file without rasterising your text.
- What resolution do the pages end up at?
- Whatever the measurement says is needed. The first pass renders at 150 dpi and weighs the result; the resolution required for 200 KB is then derived from it, since encoded size scales with pixel count and pixel count with resolution squared. Typical scanned forms land somewhere between 90 and 130 dpi, which stays readable on screen.
- How many pages fit under 200 KB?
- Roughly five to eight for a plain scanned document, because each page carries a floor cost of about 20 to 40 kilobytes once it is an image. Beyond that the arithmetic stops working, and the tool reports the smallest size it managed rather than pretending otherwise. Splitting the file is the realistic answer.
- Is the document uploaded anywhere?
- No. Rendering, encoding and measuring all happen inside your browser tab, using the same PDF engine that draws the preview. Nothing is transmitted — worth knowing when the file is a signed contract or an identity document, and also why a long scan takes a moment on an older machine.
About this tool
Before compressing anything it is worth knowing which kind of PDF you have, because the two kinds fail for opposite reasons. A text PDF stores characters and vector shapes; it is normally tiny, and when it is not, the cause is an image someone dropped in at full resolution. A scanned PDF stores a photograph of every page, which is inherently large and is exactly what this tool is built to shrink.
For scans, the process is a measured search rather than a fixed preset. The first pass renders at 150 dpi and weighs the encoded result, and because file size tracks pixel count — which is proportional to resolution squared — the resolution needed for 200 KB can be calculated from that one measurement. Most documents converge within two or three passes, and the smallest passing result is kept.
Two hundred kilobytes accommodates roughly five to eight scanned pages. The limit is not the setting but the floor cost of a page: below about 20 to 40 kilobytes an imaged page stops being readable, so a long document simply has a minimum size. When the tool hits that wall it reports the smallest size it reached and the resolution it used, so you can decide whether to split the file or drop pages rather than resubmitting blindly.
If your PDF turns out to be a text document, consider the alternatives before rasterising it. Re-exporting from the original application at lower image quality, replacing an embedded photograph with a compressed copy, or removing pages all reduce size while leaving the text selectable and searchable. Rasterising is the right tool for scans and the last resort for everything else — and because it runs locally, your document never leaves the machine either way.