Why is my PDF so large, and how do you shrink it?

Your PDF is probably full of photos, even if it looks like it is full of text.

That is the short answer. A PDF made of real text is tiny. A few pages of writing is barely any data. So when a short document balloons to twenty or thirty megabytes and an upload form bounces it, the words are not the problem. Almost always, it is images.

TL;DR: Text in a PDF is tiny. Images are not. A scanned PDF stores a full photo of every page, and high DPI plus full color multiplies that fast. To shrink it, you downsample and re-compress those images. If the file is already text, there is not much to shrink.

First: is your PDF text, or is it pictures of text?

This is the one thing worth checking before anything else, because the fix is completely different depending on the answer.

Open the file and try to highlight a sentence with your cursor. If a selection box wraps the actual words and you can copy them, the page is text. If your cursor grabs the whole page as one block, or grabs nothing, you are looking at an image. A picture of a page.

Another way: search for a word you know is on the page (Ctrl+F or Cmd+F). Real text is searchable. A plain scan is not, because to the computer it is a photograph, and there are no letters in it to find.

I do this check first every time someone hands me a “huge PDF.” Nine times out of ten the cursor slides right over the page without catching a single word, and I know before opening any tool that I am dealing with a scan.

Why scanned PDFs get so big

A scanner does not read text. It takes a photo. Every page becomes a full image, and the PDF stacks those images back to back. Ten scanned pages is ten photographs bundled into one file, with none of the efficiency a text document has.

That already makes scans heavier than typed documents. Two scanner settings then make it much worse: resolution (DPI) and color. Both multiply the pixel count, and the pixel count is the file size.

Here is the trap. Scanner defaults are often set high “to be safe,” so people scan everyday paperwork at 600 DPI in full color when none of that detail matters. The file ends up many times bigger than it needs to be for something that will only be read on a screen or printed on a basic office printer.

How DPI multiplies the size

DPI counts how many dots the scanner records per inch of paper. The cost grows with the square, not in a straight line. Double the DPI and you do not double the file. You roughly quadruple it, because you are doubling the dots in both directions at once. A page at 600 DPI carries about four times the data of the same page at 300.

Some rough numbers so this is concrete. A US Letter page is 8.5 by 11 inches. At 300 DPI that is about 2,550 by 3,300 dots, which is roughly 8 million pixels per page. At 600 DPI it is closer to 34 million. Same sheet of paper, four times the data, and none of that extra detail shows up when a human reads it.

For most documents the high setting is wasted. 300 DPI is the print standard and reads cleanly. 150 DPI is fine for anything meant to live on a screen. Scanning a tax form or a signed contract at 600 DPI gives you a giant file and no benefit anyone will ever see.

Why color scans weigh more than grayscale

A full-color scan stores three values for every pixel: one each for red, green, and blue. A grayscale scan stores a single brightness value per pixel, so it carries roughly a third of the raw data for the same page. For black text on white paper, color records nothing useful and roughly triples the weight.

If your scan is a printed letter, a receipt, or a form filled in with a pen, grayscale keeps everything that matters. Save color for pages where color actually means something: a highlighted document, a photo, a chart that needs its hues to make sense.

A worked example

Say you scan a ten-page contract in color at 600 DPI, the kind of default a lot of office scanners ship with. It is normal for that to land around 20 MB, sometimes more. Every page is a large color photograph, and there are ten of them.

Now picture the same ten pages as a document someone typed and exported as real text. The characters take a few kilobytes per page. The whole thing is well under 1 MB, often a tenth of that. Same words on the same pages, and the file is more than twenty times smaller.

That gap is the entire story of PDF size. It is not the number of pages. It is whether those pages are text or photographs, and if they are photographs, how many pixels each one holds.

What bloats a PDF besides scans

PDFs that were never scanned can still get heavy, for similar reasons: the assets inside are bigger than they need to be.

Embedded fonts add some weight, because the file carries the full typeface so it renders the same on every machine. Uncompressed or lightly compressed images inside an ordinary document add a lot more. And some files hang on to edit history or extras like page thumbnails that you never see but still pay for on disk.

The sneakiest source is the “Save as PDF” or “Export to PDF” button in some apps. Those exporters are built for fidelity, not size, so they embed full-resolution images and redundant copies of fonts. The result looks identical to a lean PDF on screen and can be several times bigger. So a PDF you made yourself from a slide deck or a design tool can be as bloated as a scan, even though every page is “real.” Somewhere in it, images are stored at a higher resolution or lower compression than the document needs.

How to actually shrink a PDF

Go after the images, because that is where the weight is.

The most effective move is downsampling: reducing the images inside the file to a sensible DPI. Dropping a 600 DPI scan to 150 for screen reading, or 300 for print, can cut most of the file while leaving it perfectly readable. The second move is re-compressing those images, usually as JPG, which throws away detail your eye barely notices in exchange for a much smaller file. (If you want the longer story on image compression, our guide on picking the right image format covers the tradeoffs.) Converting color scans to grayscale and stripping extras like thumbnails trims a bit more.

You can do all of this in your browser with the free PDF compressor at pdf.hivly.net. It downsamples and re-compresses the images inside the file, and because it runs in your browser, nothing is uploaded. The document never leaves your device. That matters when the PDF is a signed contract or a scan of an ID, exactly the kind of file you should not hand to a random website.

The tradeoff is worth stating plainly. More compression means a smaller file and softer images. For a document that will only be read on a screen, you can compress hard and never notice. For something headed to a printer, or a page with fine detail that has to stay legible, keep the DPI higher and accept a bigger file. Match the setting to the job instead of chasing the smallest number.

Why the size limit only bites at upload time

Most people never think about PDF size until a form rejects the file. Government portals, job application systems, and document upload boxes set hard caps, often a few megabytes, to keep their storage and bandwidth in check. Your file sat happily on your disk for months, then hit a wall the second you tried to send it.

The fix is the same whatever the form: get the images inside down to a sensible size. Re-scan in grayscale at a lower DPI if you still have the paper. Compress the existing file if you do not.

One honest note before you start clicking. If the file is already mostly text, compression will barely move it, because there were no heavy images to shrink. And if you over-compress a scan to squeeze under a cap, it can turn fuzzy and hard to read. Aim for the lowest DPI that keeps the document legible, check the result with your own eyes, and stop there.