How to Compress a PDF Without Uploading It: Step by Step

To compress a PDF, open Compress PDF in Priya LifePDF, add up to 10 files, choose a compression level such as Balanced, and click Compress. The tool shrinks the images inside the PDF and saves a smaller copy, all inside your browser, so the document is never uploaded.

Most of the weight in a large PDF comes from images: scanned pages, photographs, charts pasted in as pictures. Text and vector drawings are usually small. That is why the settings below are mostly about image resolution and image quality, and why a PDF that contains only typed text may not get much smaller whatever you choose.

How to compress a PDF with Priya LifePDF

1. Add your files. Drag PDFs onto the upload area or click to choose them. You can add up to 10 at once, and use Add more PDFs to top up the list.
2. Optionally set a target size. Under Target File Size Limit (Guidance), pick 2, 5, 10 or 20 MB, or enter a custom figure. The tool then adjusts the manual settings to aim for that size.
3. Choose a compression level. Pick Maximum, Balanced, High Quality or Near Lossless. Balanced is selected by default.
4. Check the switches. Optimize images and Strip metadata are on by default; Convert to Grayscale (B&W) is off.
5. Click Compress. Files are processed two at a time with a progress bar for each, and you can cancel at any point.
6. Review and download. Each file shows its original size, new size and percentage saved. Download files one by one (named yourfile_compressed.pdf) or, when several are done, use Download All (.zip).

If the result is not what you wanted, Re-compress with Different Settings returns you to the settings with the same files loaded.

What the compression settings actually do

The standard engine re-saves the PDF structure more compactly and, when Optimize images is on, downscales and re-encodes images as JPEG. Each level sets a target resolution and a JPEG quality:

• Maximum: 120 DPI at 70% JPEG quality. The smallest files, with visible softening in photos and scans.
• Balanced: 150 DPI at 80% quality. A sensible default for reading on screen and emailing.
• High Quality: 200 DPI at 88% quality. Better for documents that will be printed.
• Near Lossless: 300 DPI at 94% quality. Keeps fine detail, with the smallest saving.

Manual optimisation settings let you override the level with your own values: a target DPI from 50 to 300 and a JPEG quality from 10% to 100%. Choosing a target file size switches these manual settings on and fills them in for you; for a large reduction it may also switch greyscale on, so check the toggles before you compress.

Strip metadata removes the document information dictionary (author, title, keywords and similar fields) and the embedded XMP metadata. Convert to Grayscale turns the images that are re-encoded into greyscale, which saves space on colour scans.

Some content is deliberately left alone. JPEG images already at or below the target resolution are not re-encoded, and images in CMYK or other complex colour spaces are skipped, so print-ready files may shrink less than you expect.

Fallback engines

If the standard engine cannot process a file, the tool tries a second engine (PyMuPDF) and then a third that rasterises the pages. The result row shows via PyMuPDF or via Rasterise when this happens. Rasterising turns every page into a JPEG image, so the text is no longer selectable or searchable and links stop working. If you see that label on a document that needs searchable text, keep the original instead.

Tips and common mistakes

• Start with Balanced, then adjust. Open the result and zoom in on the smallest text in a scanned area before you move to Maximum.
• Scans with handwriting need care. Signatures, initials and handwritten figures on a scan are images. Use High Quality or Near Lossless and check them at 100% zoom.
• Check the percentage. A text-only PDF may barely change. If a file did not get meaningfully smaller, keep the original rather than sending a re-saved copy.
• Mind the metadata switch. Strip metadata is on by default. Turn it off if the title or author fields matter to the recipient; use Remove Metadata separately if you want to do that step deliberately.
• Password-protected files. Remove the password first with Decrypt PDF (you need to know it), then compress.
• Too big even after compression? Split it with Split PDF and send it in parts, or convert the whole document to greyscale with PDF to Greyscale if colour is not needed.
• Keep the original. Compression is lossy for images. Treat the compressed file as a copy for sending, not a replacement.

Why compress locally instead of uploading?

Files you want to compress are often the large ones: scanned contracts, signed forms, records full of photographs. On many online compressors the whole file is sent to a server first. Compress PDF works in the browser tab using JavaScript, with a WebAssembly build of PyMuPDF loaded from the site only if the fallback engine is needed, so the document itself stays on your device. You do not have to take that on trust: our guide on how to check whether a PDF tool uploads your files shows how to confirm it in your browser's Network tab. For the wider picture, see is it safe to upload PDFs to online converters?

Example: sending a scanned training record pack to a partner site

A quality associate needs to email a pack of scanned training records and a signed SOP acknowledgement to a contract manufacturing partner, and the combined scan is too large for the mail server. The pack contains names and signatures, so uploading it to an unknown service is not attractive.

1. Open the scan in Compress PDF and choose High Quality, because the signatures and handwritten dates must stay legible.
2. Leave Strip metadata on, since the recipient only needs the pages.
3. Compare the result with the original at 100% zoom, page by page, and confirm the engine label shows no fallback.
4. Send the compressed copy and keep the original scan as the retained record.

The compressed PDF is a new file, so it should be handled under your own document control procedures like any other copy. Priya LifePDF is a general-purpose tool, not a validated GxP system; our article on online PDF tools in pharma and GMP work covers what to consider before using any PDF tool on regulated records.