Why is my PDF so large?

Almost always high-resolution images (such as phone-camera scans), duplicated or unused embedded fonts, and leftover objects from old revisions. The text itself is rarely the cause.

Will compressing a PDF make the text blurry?

It shouldn't. Good compression down-samples images but leaves real text as selectable text. Blur happens only when a poor tool rasterizes the whole page into a low-quality JPEG.

What DPI should I compress to?

About 150 DPI is plenty for on-screen reading and email. Use 300 DPI if the document will be printed.

How do I compress a PDF to 100KB or under 1MB?

Target the images, since they're almost always the cause of size. Down-sample scans and photos to a lower DPI and strip embedded thumbnails and metadata — that alone usually gets an image-heavy file under 1MB. A text-only PDF is typically already that small; if it isn't, an unused embedded font is the likely culprit.

optimize →

How to compress a PDF without turning the text into mush

MSMaya SundaramMay 2, 20267 min read

Arthize guide cover — how to compress a PDF

The short version

PDFs are big because of high-res images, duplicated fonts, and leftover junk — not because of text.
Good compression down-samples images, subsets fonts, and strips dead objects while keeping text selectable.
Choose a target by intent: 150 DPI for screen/email, 300 DPI for print.
Never compress the same file twice; if it's big because it's long, split it instead.

There's a special frustration in watching a 38 MB PDF bounce off a mailbox that allows 25, then running it through a "compress" button and getting back a 36 MB file that now also looks like it was faxed underwater. Compression isn't magic, but it isn't a coin flip either. Once you know what's making a PDF big, you can make it small and keep it readable.

Why your PDF is huge in the first place

Text is tiny. A 200-page text document is often under a megabyte. So when a PDF is enormous, the weight is almost always one of three things:

High-resolution images. A phone-camera scan at 4000 px per page is the usual culprit. Five of those and you're at 30 MB.
Embedded fonts and duplicates. Documents assembled from many sources often embed the same font five times, or carry fonts no page actually uses.
Leftover junk. Old revisions, unused objects, and metadata that never got cleaned up.

What good compression actually does

Real PDF compression works on those three problems directly: it down-samples images to a sensible resolution (150–200 DPI is plenty for screen reading; 300 if it'll be printed), de-duplicates and subsets fonts so only the glyphs you use are stored, and strips dead objects. Crucially, it leaves your real text as text. You can still select and search it afterward.

How to compress a PDF the right way

Open the compress tool and load the file.
Pick a target by intent, not a random slider: screen / email (aggressive, ~150 DPI), balanced, or print (gentle, keeps ~300 DPI).
Compress, then actually open the result and zoom to 200% on a page with small text.
If it's still too big, don't compress twice (that just degrades it) — try a lower DPI target once, or split the document.

Don't compress the same file twice

Each lossy pass throws away detail it can never get back. Compressing an already-compressed file gives you diminishing returns and accelerating ugliness. If one pass at "email" quality isn't enough, the file is probably a candidate for a different approach entirely.

When compression is the wrong tool

The file is big because it's long

If your PDF is large because it genuinely has 400 pages, compressing helps only so much. Splitting it into the part the recipient needs is faster and cleaner — and they didn't want all 400 pages anyway.

You're about to merge anyway

If you're combining several files, merge first, then compress the final document once. Compressing each input and then merging stacks up quality loss for no benefit.

It's a scan you need to read or search

A heavy scanned PDF often compresses better and becomes useful if you OCR it first — you get searchable text and the option to keep the page images lighter.

A note on where it happens

Compression reads every page of your document — including whatever's on them. Browser-based "free PDF compressor" sites upload the full file to their servers to do the work. For a public brochure, who cares; for an invoice with bank details, you've just handed a copy to a server you can't see. We run compression inside the same private workspace as the rest of Arthize for exactly that reason.

Compression is the "optimize" leg of the broader PDF workflow guide — usually the last step before a file goes out the door.

Frequently asked

Why is my PDF so large?: Almost always high-resolution images (such as phone-camera scans), duplicated or unused embedded fonts, and leftover objects from old revisions. The text itself is rarely the cause.
Will compressing a PDF make the text blurry?: It shouldn't. Good compression down-samples images but leaves real text as selectable text. Blur happens only when a poor tool rasterizes the whole page into a low-quality JPEG.
What DPI should I compress to?: About 150 DPI is plenty for on-screen reading and email. Use 300 DPI if the document will be printed.
How do I compress a PDF to 100KB or under 1MB?: Target the images, since they're almost always the cause of size. Down-sample scans and photos to a lower DPI and strip embedded thumbnails and metadata — that alone usually gets an image-heavy file under 1MB. A text-only PDF is typically already that small; if it isn't, an unused embedded font is the likely culprit.

Maya Sundaram

Co-founder & document-tooling engineer, Arthize

Maya has spent the last decade building document-processing systems — first for a legal-tech startup that ingested millions of scanned filings, now at Arthize where she owns the conversion, OCR and compression pipelines. She has opinions about Ghostscript flags.