Why are some PDFs so large? A guide to PDF file size

Two PDFs with the same number of pages can differ in size by a factor of a hundred. A ten-page report exported from a word processor might be 200 KB, while ten scanned pages from a phone can easily reach 30 MB. Knowing where the weight comes from makes it much easier to bring a file under an email or upload limit.

1. Images, almost always

Pictures are the biggest cause of large PDFs. A scanned page is one large photograph of paper. At 300 dots per inch, an A4 page is roughly 2,480 × 3,508 pixels, nearly nine million pixels, and a colour scan stores three values for every one of them. Even compressed, that adds up quickly across many pages.

Phone cameras make it worse. A modern phone takes 12 to 50 megapixel photos, far more detail than a page of text needs. Put twenty such photos into a PDF and you have a file that’s slow to send for no visible benefit.

2. Embedded fonts

To look identical everywhere, PDFs carry their fonts. Usually only the characters that are used are embedded, which is small. But some software embeds entire font families, and fonts for Chinese, Japanese or Korean can be several megabytes each. A document can be large while containing only text.

3. Leftovers from editing

PDFs can be updated by appending changes to the end of the file instead of rewriting it. Each save may leave the previous version of an edited page inside. Merging documents can also duplicate the same logo or font several times. These leftovers are invisible but still count towards the size.

4. Extras you can’t see

Attachments, embedded page thumbnails, detailed metadata, colour profiles and form scripts all add weight. They are rarely the main cause, but in large batches they matter.

What compression can and can’t do

Good PDF compression attacks these causes in order of impact:

• Resampling images: reducing a 4,000-pixel-wide photo to around 2,000 pixels keeps it sharp on screen and in normal printing while cutting its size by about three-quarters.
• Re-encoding images: saving them as JPEG at a sensible quality, which suits photographs and scans.
• Cleaning the structure: removing unused objects and duplicates, compressing any data stored uncompressed, and packing objects together efficiently.

Compression can’t safely remove fonts the pages actually use, and it can’t make text-only documents much smaller, because they were small to begin with. If a 300 KB report only shrinks to 280 KB, that’s expected: it was already efficient.

Typical limits to aim for

Many email services cap attachments at around 20 to 25 MB per message, including all attachments together. Online forms for job applications, visa applications and university admissions often set lower limits of 2, 5 or 10 MB per file. Check the exact figure before you start, because it tells you how aggressive to be.

A practical order of attack

1. Try Compress PDF at the Recommended level. For scanned documents this alone often cuts the size by 70% or more.
2. If that isn’t enough, use the Strong level. Check the result on screen before sending.
3. Remove pages that don’t need to be there with Delete Pages.
4. Colour isn’t always needed. Converting a scan to greyscale can reduce it substantially.
5. If the limit is per file, split the document into parts.
6. For scans you only need to read, Maximum compression turns each page into a compact image. Text won’t be selectable afterwards, so keep the original.

Preventing large files in the first place

When scanning, 200 to 300 DPI is plenty for documents, and black-and-white or greyscale mode is enough for text. When exporting from office software, look for an option such as “Optimise for: Minimum size” or “Reduce file size”. And when photographing documents with a phone, use the built-in document scanner in apps such as Notes on iPhone or Google Drive on Android: they crop and flatten the page and produce far smaller files than ordinary photos.