Most large PDFs are large because of images: scanned pages and photos stored at a higher resolution than any screen needs. To reduce PDF file size without visible quality loss, first try lossless optimisation, which keeps text and links intact, then downsample images to around 150 DPI if the file is still too big. Re-rendering every page as an image gives the biggest savings on scans but removes selectable text, so keep the original.
- Images, especially scans and phone photos, account for most of the size of a typical large PDF.
- Lossless compression is always safe but often saves little; image compression saves much more but changes the images.
- Around 150 DPI is sharp on screen; 300 DPI is only needed for printing or fine detail.
- Many email providers cap attachments at around 20 to 25 MB, and upload portals are often much stricter.
- Always check the result for readability, text selection and links before sending it.
A text-only PDF of dozens of pages can be smaller than a single phone photo, while a short scanned form can run to many megabytes. Knowing what is inside the file tells you which technique will work.
Why PDFs get so large
- High-resolution images. A modern phone camera produces photos of 12 megapixels or more. Inserted into a document, the full image is usually kept even when it is displayed a few centimetres wide.
- Scanned pages. A scan is a photograph of each page. Scanners set to 300 or 600 DPI in full colour produce large images even for a page of plain black text.
- Inefficient image encoding. Some tools store photos with lossless compression instead of JPEG, which can make them several times larger.
- Embedded fonts. PDFs embed fonts so they look the same everywhere. Most tools include only the characters used (a subset), but some embed complete font files, which matters most for large Asian-language fonts.
- Duplicated resources. Merging PDFs often leaves several copies of the same fonts and logos, one from each source file.
- Leftover data. Editors that save incrementally append changes to the end of the file, keeping old versions inside it. Metadata, page thumbnails, embedded attachments, hidden layers and form scripts add more.
Lossless optimisation vs image compression
There are three levels of PDF compression, and the trade-offs between them are very different.
| Method | What changes | Text, links and forms | Typical saving | Best for |
|---|---|---|---|---|
| Lossless optimisation | Removes unused objects and old revisions, packs objects and streams more efficiently, merges duplicates | Kept exactly | Small to moderate; little on already-optimised files | Digital documents, contracts, forms |
| Image downsampling | Embedded images are reduced in resolution and recompressed as JPEG; text stays as text | Kept | Large on photo-heavy documents | Reports and brochures with photos |
| Page re-rendering | Each page is drawn as one image and a new PDF is built from those images | Lost: text cannot be selected or searched, links stop working | Large on scans; can make text-only files bigger | Scans and files that must fit a strict limit |
Our Compress PDF tool offers a Lossless mode and two re-rendering modes: Recommended draws each page as a JPEG at 150 DPI, and Strong does the same at 100 DPI. If the result would be bigger than the original, which can happen with short text-only files, it keeps your original. It runs in the browser, so confidential documents are not uploaded.
Re-rendering a digital document is usually a poor trade: a page of vector text is very compact, whereas a picture of that page is not, and you lose search, copy and accessibility for screen readers. Reserve re-rendering for scanned documents, which were images anyway.
Choosing the right resolution
DPI (dots per inch) describes how many pixels are used for each inch of the printed page. An A4 page is about 8.3 by 11.7 inches, so:
| Resolution | A4 page in pixels (approx.) | Suitable for |
|---|---|---|
| 72–100 DPI | 600–830 wide | Previews; small print may look soft |
| 150 DPI | 1240 × 1754 | Reading on screen, email, most uploads |
| 200–300 DPI | 1650–2480 wide | Printing, fine print, OCR, signatures |
| 600 DPI | 4960 × 7016 | Archival scans, detailed drawings |
Doubling the resolution quadruples the number of pixels, which is why dropping a 300 DPI scan to 150 DPI often cuts its image data dramatically while remaining perfectly readable on screen.
Size targets for email and uploads
Many email providers cap attachments at around 20 to 25 MB per message, and some corporate mail servers set lower limits. Attachments are also encoded for transport, which increases their size by roughly a third, so a file that is just under the stated limit may still be rejected. As a practical rule:
- Email: aim for well under 10 MB where you can; anything over about 15 MB risks bouncing.
- Job applications, government and university portals: check the stated limit, which is often only a few megabytes per file.
- Messaging apps and web forms: limits vary widely; smaller files also open faster on phones.
When a file is simply too big to email even after compression, a cloud storage link is a better option than degrading the document.
How to compress a PDF step by step
- Open Compress PDF and add the file.
- Choose Lossless first and check the new size.
- If it is still too large and the document is a scan or mostly photos, try Recommended (150 DPI).
- Use Strong (100 DPI) only when you must meet a hard limit, and check that small print is still legible.
- Download the result and keep the original file unchanged.
Desktop alternatives work on the same principles. In Microsoft Word for Windows, File > Save As with the PDF type offers an Optimize for: Minimum size (publishing online) option. Adobe Acrobat Pro has Reduce File Size and a more detailed PDF Optimizer. On Linux and macOS, Ghostscript can rewrite a PDF with image downsampling while keeping the text:
gs -sDEVICE=pdfwrite -dPDFSETTINGS=/ebook -dNOPAUSE -dBATCH -o smaller.pdf original.pdfThe /ebook preset targets about 150 DPI; /screen goes lower and /printer keeps 300 DPI.
Split the file or remove pages
Sometimes the best way to make a PDF smaller is to send less of it. If only a few pages are needed, delete the rest with Remove PDF Pages. If every page is needed but the whole file exceeds an attachment limit, use Split PDF to divide it into parts and send them in separate messages, numbered so the recipient can keep them in order. Blank pages and separator sheets from a scanner are worth removing too.
Archiving a PDF into a ZIP rarely helps, because PDF content is already compressed internally; our guide to archive formats explains why.
Making scanned documents smaller
The cheapest saving happens at scan time. For a typical text document, scan at 200 or 300 DPI rather than 600, and use greyscale or black and white instead of colour unless colour carries meaning, such as a stamp or a highlighted signature. Phone scanning apps that crop and flatten pages also produce much smaller files than ordinary photos.
For photos taken with a phone, you can control the result more precisely by working on the images yourself:
- If the pages are already in a PDF, convert them to images with PDF to JPG at 150 DPI.
- Shrink the images with Compress Image, adjusting quality until text is still crisp.
- Rebuild the document with JPG to PDF, choosing A4 or US Letter so the pages print at the right size.
Note that, like page re-rendering, this produces image-only pages. If the recipient needs searchable text, run the scan through OCR (optical character recognition) software afterwards.
Check the result before sending
- Read the smallest text at 100% and 200% zoom, including footnotes, tables and figures.
- Check signatures, stamps and photos for blur or JPEG artefacts.
- Try selecting and searching text if the recipient will need to copy from it.
- Click a link and test form fields in digital documents.
- Compare the page count with the original.
- Print one page if the document is meant to be printed.
Frequently asked questions
Can I reduce PDF size without losing any quality?
Yes, with lossless optimisation, which removes waste without touching the content. The saving depends on how efficiently the file was created and may be small. Larger reductions require compressing images, which is lossy, although at 150 DPI the difference is usually invisible on screen.
Why is my scanned PDF so large?
Each page is a full image, often scanned at high resolution in colour. Rescanning at 200 to 300 DPI in greyscale, or compressing the existing file to 150 DPI, usually reduces it substantially.
Why can't I select text after compressing my PDF?
The compressor re-rendered every page as an image, which removes the text layer. Use a lossless or image-downsampling method instead, or keep the original for copying text.
Why did compression make my PDF bigger?
A text-only PDF is already compact, and turning its pages into images adds data. Our tool keeps the original in that case; for such files, lossless optimisation is the only useful option.
How do I get a PDF under 2 MB for an upload portal?
Try lossless first, then 150 DPI, then 100 DPI. If it is still too big, remove pages that are not required, or split it into several files if the portal accepts more than one.