Why Is My PDF So Large? Causes and Fixes That Work

Updated 30 October 2026 · by the My Gadgets team

A PDF is almost always large for one reason: it contains big pictures. Scanned pages, phone photos of documents and high-resolution images account for nearly all the PDFs that bounce off an upload limit. Text, by comparison, costs almost nothing — a 50-page report with no images is usually well under 1 MB. So the fix that matters is making the pictures inside the file smaller — which Compress PDF does in one step — and the rest of this article shows how to tell whether that is your problem and how to do it without turning your document into mush.

How big should a PDF be?

A rough sense of normal makes it obvious when something is wrong:

If your 4-page file is 18 MB, you are in the last two rows: every page is a large image.

What is making my PDF large?

1. The pages are images, not text

This is the cause nine times out of ten. A scanner app, a phone camera or a "print to PDF" of a photo produces pages that are pictures of text. Try selecting a word on the page: if you can't, the page is an image, and its file size is set by the image's pixel count and how it was compressed.

Phone cameras make this worse. A modern phone photo is 12 megapixels or more — about 4000 × 3000 pixels — which is far more detail than a printed A4 page needs. Around 1600 to 2000 pixels on the long side is plenty to read and print a page of text.

2. Images were stored without compression

The same page can be stored as a PNG (lossless, large) or a JPEG (lossy, small). For a scanned page, PNG can easily be five to ten times bigger with no visible benefit. Some tools convert every image to PNG when building a PDF, which quietly multiplies the size.

3. Leftovers from repeated editing

Some editors save changes by appending them to the end of the file rather than rewriting it, so an often-edited PDF can carry old versions of pages it no longer shows. This is far less common than causes 1 and 2, and the saving from cleaning it up is usually modest.

4. Embedded fonts

PDFs embed the fonts they use so the document looks the same everywhere. A full font for a script with many characters can add a few megabytes. It is rarely the main problem, and removing fonts is not something you should do by hand — the document may stop displaying correctly.

Which fix should I use?

First try: Compress PDF

Open the file in Compress PDF and run it on Standard with image quality Medium. It re-encodes every large image inside the PDF as a JPEG and scales it so its longer side is at most 2400 pixels — still sharp for an A4 page — while leaving text, fonts and vector drawings exactly as they were. A scanned or photographed PDF often shrinks by 70–90 per cent this way, and any selectable text it has stays selectable.

If it is still too large, run the original again with Aggressive (1800 pixels) or Maximum (1300 pixels) and quality Low. Always start from the original rather than compressing the compressed copy, and check one page at full zoom before sending.

The tool tells you how many images it re-encoded. If it says few or none and the file is still big, the images are in a format it leaves alone — CMYK colour from professional scanners, or images with special masks — and the next route handles those.

If Compress PDF can't shrink the images: rebuild the pages

When images are in a format the normal method leaves alone, Compress PDF says so in its result and suggests the other method. Choose Method → Rebuild every page as an image and run it again on the original. It redraws each page as a fresh JPEG at the level's size (2400 pixels on the longer side at Standard, 1800 at Aggressive, 1300 at Maximum), so it works on anything a PDF reader can show, and keeps the page sizes.

The trade-off: the rebuilt PDF is a set of pictures, so any selectable text becomes part of the image. That costs nothing for a plain scan, but keep the first method for documents whose text you want to stay searchable.

If the text is selectable and the file is still large: send less of it

For a real text PDF, the size often comes from the number of pages or a few heavy pages:

• Remove pages nobody needs with Remove Pages from PDF. A cover sheet that is a full-page photo can be the heaviest page in the file.
• Send only the part that is needed with Split PDF — an application usually wants the signed page, not the whole 40-page contract.

If you control how the PDF is made: fix it at the source

The best fix is never creating the large file:

• In a scanner app, choose 200–300 dpi and "greyscale" or "black and white" for text pages. Colour at 600 dpi is for photographs, not letters.
• On a phone, use the document scanner built into the Notes app (iPhone) or Google Drive (Android) rather than the camera; they crop, flatten and compress the page for you.
• When exporting from Word or Google Docs, choose the standard or "minimum size" PDF option and keep pictures at their displayed size, not their original camera size.

What about "online compressors"?

Most online PDF compressors do what Compress PDF does: they re-encode the images inside the file at a lower resolution. That works, but it means uploading the document to someone else's server. For a passport scan, bank statement or signed contract — exactly the files that run into upload limits — the tools here do the same job inside your browser, and the file never leaves your device.

Quick answers

How do I get a PDF under 2 MB? Run Compress PDF on Standard first; scanned pages usually end up at a few hundred kilobytes each. For a tight limit on a long scan, use Aggressive or Maximum from the original, or split the file if the portal accepts several uploads.

Why did Compress PDF barely change my file? Either the PDF is mostly text — already small, with little to compress — or its images use a colour format the tool leaves alone. In the second case, Compress PDF tells you so; switch its method to Rebuild every page as an image.

Will compressing make my PDF blurry? Re-encoding images always loses some detail. On Standard with Medium quality, text on a scanned page stays sharp; softness usually appears first around stamps, photos and thin lines at the strongest settings.

Does compressing remove the text from my PDF? No. Compress PDF changes only the images; text, fonts and links are kept. Only the Rebuild every page as an image method turns pages into pictures.

What to do next

Start with Compress PDF. If it can't shrink the images, run it again with Rebuild every page as an image. For long text documents, send only the pages that are needed. If you are preparing several documents for an application, the step-by-step guide to preparing documents for an online application covers the whole process, from photographing a page to signing the form.