How to Compress a PDF

Most advice on this points you at an upload site. Before you send a document containing who-knows-what to a stranger's server, it is worth knowing what actually makes a PDF large and which fixes are available offline.

First, find out why it is large

PDF size comes from one of three things, and the fix is completely different for each:

Cause	Typical size	Tell-tale sign
Images or scans	5–100 MB	Pages look like photographs; text cannot be selected
Embedded fonts	1–5 MB	Text-only document that is still surprisingly big
Accumulated revision history	varies	File grew each time it was edited and saved

PDF is an open ISO standard, and the reason a text PDF and a scanned PDF can both be "a PDF" while behaving completely differently is that the format can hold either vector text or page images — the PDF Association maintains the specification and its sub-standards.

The quick test: try to select text on the page. If you can, it is a real text PDF and it should not be large — the problem is fonts or revision bloat, both of which re-printing fixes. If you cannot select any text, every page is an image, and only image compression will help.

Method 1: print to PDF (Windows, no software)

This is the one most people do not know about, and it is built in.

1. Open the PDF in any viewer, including a browser.
2. Press Ctrl + P.
3. Choose Microsoft Print to PDF as the printer.
4. Print, and save under a new name.

What happens is that the document is re-rendered from scratch, so incremental save history, unused objects and duplicated font data are discarded. On text-heavy documents this often cuts size substantially.

Two things it costs you. The output usually loses selectable text if the source was already an image, and it flattens interactive elements — form fields, bookmarks and internal links stop working. For a document you are sending someone to read, that is fine. For a fillable form, it is not.

Keep the original. Always save the compressed version under a new name.

Method 2: compress the images before they become a PDF

If the PDF is made from photographs or scans, the images are the entire file. Fixing it at the source is far more effective than compressing afterwards.

If you are scanning: scan at 200–300 DPI, not 600. For a document that will be read on screen or printed normally, 300 DPI is plenty and 600 DPI is four times the data for no visible benefit. Choose black and white or greyscale for text documents rather than colour — a greyscale scan of a printed page is a fraction of the size and looks identical.

If you are photographing documents: most phone scanner apps have a document mode that converts to high-contrast black and white. That is not a cosmetic filter — it removes the photographic noise that compresses badly, and can reduce size by an order of magnitude.

If you already have the images: resize them before assembling. A 4000-pixel-wide photo of an A4 page is far more resolution than the page contains; 1500–2000 pixels wide is ample.

Method 3: split it instead

Often the real requirement is not "smaller file" but "under the 10 MB limit on this form". Splitting a long document into parts sidesteps compression entirely and loses no quality.

In most viewers, print to PDF with a page range — 1-20, then 21-40 — to produce separate files. Worth checking whether the recipient accepts multiple attachments before degrading image quality to fit one.

Method 4: online compressors, and when not to use one

Online tools genuinely work well, and for a non-sensitive document they are the easiest option. But uploading a file sends it to someone else's server.

Do not upload:

• Anything with bank details, ID documents or passport scans
• Medical records
• Signed contracts or anything with a handwritten signature
• Client or employer material covered by confidentiality
• Anything containing other people's personal data

A signature image lifted from a PDF is directly reusable, which makes signed documents a particularly bad thing to hand over. For anything in that list, use the print-to-PDF method — it is offline, and nothing leaves your machine.

If you do use an online tool, prefer one that states how long files are retained, and delete the file afterwards if the option exists.

What does not work

• Zipping a PDF. PDFs are already internally compressed, so a ZIP typically saves a few percent at most. It also makes the file more annoying to open.
• Renaming or changing the extension. Does nothing to size.
• Compressing repeatedly. Image compression is lossy; each pass throws away more detail. Two passes produce a visibly worse document, not a much smaller one.
• Removing metadata. Real for privacy, negligible for size — metadata is kilobytes.

Choosing a target size

Purpose	Reasonable target	Approach
Email attachment	Under 10 MB	Print to PDF; split if needed
Online form upload	Often 2–5 MB	Reduce image resolution
Reading on screen	150 DPI images	Greyscale where possible
Printing at home	300 DPI images	Do not compress further
Archiving	Keep the original	Compress a copy only

The last row is the one worth taking seriously. Compression is lossy and irreversible — once detail is discarded it cannot be recovered. Keep the full-quality original somewhere and treat compressed versions as disposable copies made for a specific purpose.

If you are unsure how the numbers translate, our guide to MB, GB and TB explained covers why a "10 MB limit" may not mean quite what you expect, and the percentage calculator will tell you how much you actually saved.

Understanding the numbers helps: MB, GB and TB explained covers what those file sizes actually mean, and speeding up a website applies the same compression thinking to pages.

Frequently asked questions

How do I compress a PDF without losing quality?

Print it to PDF using the built-in Microsoft Print to PDF printer. That rebuilds the file and discards revision history and duplicated font data without re-compressing images, so text documents shrink with no visible loss. Any method that reduces image resolution will lose some quality by definition.

Why is my PDF so large?

Almost always images. Try selecting text on a page: if you cannot, the page is a picture and the images are the file size. Scans and phone photos at high resolution produce very large PDFs. Text-only documents that are still large usually carry embedded fonts or accumulated save history.

Is it safe to use an online PDF compressor?

For non-sensitive documents, generally yes. Do not upload anything with bank details, identity documents, medical records or signatures — you are sending the file to someone else's server. Use the offline print-to-PDF method for those.

Does zipping a PDF make it smaller?

Barely. PDFs are already compressed internally, so a ZIP usually saves only a few percent while making the file harder to open. It is not worth doing for size alone.

What DPI should I scan documents at?

200–300 DPI for normal text documents. 600 DPI produces roughly four times the data with no benefit for on-screen reading or ordinary printing. Scan in greyscale or black and white rather than colour unless the colour carries information.

Will compressing a PDF break the links or form fields?

Printing to PDF will — it flattens the document, so form fields, bookmarks and internal links stop working. If you need those preserved, use a tool that optimises images while keeping the document structure intact.

Conclusion

Check whether you can select the text. That single test tells you whether you are dealing with a text document that print-to-PDF will fix in thirty seconds, or an image-based file where resolution is the only lever.

And keep the original. Compression is one-way, and "smaller" is only ever worth having relative to a specific limit you are trying to meet.