K-magpie › Kmagpie Tools › Compress PDF
Compress PDF to a Target Size
Shrink a scanned PDF to fit a size limit. Pick a target like "under 5MB" and the tool finds the resolution and quality for you, with a before/after preview so you can check the result with your own eyes. Your file never leaves your browser — no upload, no sign-up.
| Original → result size | |
|---|---|
| Reduction | |
| Settings applied | |
| Pages | |
| Text selection & search | No longer possible — every page is rebuilt as an image |
Compresses a scanned PDF to a target size such as 2, 5 or 10MB by automatically searching for the best resolution and quality. Your file is never uploaded — everything runs inside your browser — and you can compare before/after quality on screen. Because pages are re-rendered as images, text selection and search are lost, so keep your original file.
Try the Magpie TestsYour birth date is the only input and it takes 30 seconds — zombie survival rate, past life, idol position, guardian spirit. Made to compare with friends.
→
Why compress to a target size
People usually shrink a PDF because they have to submit it somewhere: an email attachment limit, a government portal's "10MB max", a job application's "5MB max". What matters then isn't the percentage saved — it's whether the file gets under the limit. So instead of asking you to pick a compression strength, this tool asks for the target size first, then binary-searches for the resolution and quality combination that damages the image least while still fitting. For long documents it first approximates with a few representative pages before encoding everything in one pass, which keeps the number of retries down.
Scanned PDFs and text PDFs compress differently
A PDF made by scanning paper or photographing it with a phone stores each page as one big picture. Pictures shrink dramatically when you lower the resolution and adjust JPEG quality, so these files respond well to the re-rendering approach this tool uses — redrawing each page as an image and assembling a new PDF from those images.
A PDF exported straight from Word or Google Docs is different: its text is stored as fonts and coordinates, which is tiny to begin with. Re-render a file like that and a few dozen kilobytes of text data become a full-page image — the file can actually get bigger, and text selection, search and copying disappear. That's why this tool analyses the first pages the moment you open a file, checking the amount of real text and the file size per page to judge whether it's a scan or a text document, and warns you clearly before proceeding if it's the latter. The evidence behind the verdict is shown on screen too.
The trade-off: text selection and search are lost
After re-rendering, every page of the PDF is an image. You can't drag to select text in a viewer, in-document search finds nothing, and text can't be copied or read aloud by a screen reader. If the original was a scan, those features never existed and you lose nothing; if it was a text PDF, this is a real loss. Either way, keep the original file and treat the compressed copy as a submission-only version.
Why phone photos of documents are so large
Modern phone cameras shoot at 12 megapixels or more, so a single photo is around 4,000×3,000 pixels — higher resolution than a Letter or A4 page scanned at 300dpi, when roughly 150dpi is plenty for reading a document. In other words, a photographed document that weighs 3–5MB per page is almost always a case of resolution overkill. Re-rendering trims that excess, so the file shrinks a lot without hurting legibility. After compressing, zoom into the before/after preview and make sure the text is still crisp.
When compression won't help
A file that was already produced at low resolution with strong compression has almost nothing left to give. Feed in a scan that has been compressed once before and the saving may be tiny — or re-encoding losses may even make it bigger. If the result comes out larger than the original, just use the original. Text-heavy PDFs often won't shrink this way either, for the reasons above. And if you set the target too small, even the lowest settings (72dpi, low quality) may not reach it — in that case the tool shows the result along with an honest notice that the target was missed.
Troubleshooting
The file won't open
Password-protected PDFs can't be opened. If a viewer asks for a password when you open the file, you'll need an unprotected copy first. Damaged files, or files that merely have a .pdf extension, won't open either.
The result is slightly over the target
That means even the lowest settings couldn't reach the target. With many pages and a small target it can be physically impossible. Submit the document in parts, or use the PDF editor to keep only the essential pages and compress again.
The quality got too bad
The smaller the target, the more resolution and quality get cut. If your submission limit has room, raise the target one step and compress again. Your original is untouched, so you can retry as many times as you like.
Long documents are slow or fail
All rendering and encoding happens inside your browser, using your own device's CPU and memory. Files over 100 pages or 100MB can fail partway through. When that happens, splitting the document and compressing it in parts is the safe route.
The download is always named compressed.pdf
The tool saves under one fixed English filename so downloads behave identically in every browser. Rename the file after downloading if you need a specific name.