
OpenToola Guides
How to Compress a PDF Without Destroying the Useful Part
Reduce a PDF for email or a portal, choose between structural and image compression, and check that text, links and signatures still work.
A portal accepts uploads up to 5 MB. Your scanned report is 18 MB. Or an email bounces because the attachment is too large, and you need to send it today. Compression is the obvious answer, but “compress PDF” means two very different things, and one of them can turn a searchable document into a stack of pictures.
The right method depends on what is inside the PDF. A text document with a few charts behaves differently from a 40-page scan. Before choosing a setting, look at whether you need the text to remain selectable, the links to work, or a signature to stay valid.
Why PDFs get large
A PDF is a container. It can hold text, fonts, vector drawings, images, form fields, annotations, bookmarks and embedded files. Size usually comes from one of four places:
- scanned pages stored as high-resolution images;
- photos or charts placed at print resolution;
- fonts embedded several times or unused objects left by an editor;
- a document assembled from many sources without optimisation.
A PDF made from Word is often small because the text is stored as text and the images are compressed. A scanned PDF is large because every page is a photograph. Compressing the first with an image method is overkill; compressing the second with a structural method may barely change the size.
Two kinds of compression
Structural compression rebuilds the PDF’s internal objects and uses object streams. It does not rasterise the pages. Text stays text, links stay links, and the document usually looks identical. The saving depends on how wasteful the original file is. Sometimes it is a few percent; sometimes it is much more.
Image compression renders each page as a picture, then rebuilds the PDF from those images. This is the method that dramatically shrinks scanned documents. It is also the method that can ruin a text PDF: the text becomes part of the image, so it cannot be selected, searched, copied or read by a screen reader. Form fields, bookmarks and annotations may disappear.
OpenToola exposes both modes. The default is structural. Use image mode only when you have checked the document and can accept that trade-off.
Choose the mode by document type
Use structural compression for contracts, invoices, reports, forms and anything created digitally. If the file is already small, stop there. If it is still too large, look for the real cause: a high-resolution image, an unnecessary page, or a duplicate attachment. Removing the cause is better than rasterising the whole document.
Use image compression for scans where the text is already an image. A scanned receipt, an old certificate or a photographed page has no selectable text to lose. Rebuilding it as a JPEG at a sensible resolution can cut the file size by a large factor.
Do not use image compression for a PDF that must remain accessible, searchable or legally exact. A document with a digital signature should not be rasterised; the signature will no longer validate. A PDF/A archive should not be rebuilt as images either.
Step by step
Open the Compress PDF tool. Select the PDF. Check the original size shown by the tool.
Start with structural mode. Click Compress PDF and compare the result with the original. If the file is small enough, download it and stop. There is no reason to apply a more destructive method when the safe one works.
If the file is still too large and it is a scan, switch to image mode. Set the DPI and quality. A DPI of 150 and quality around 0.7 is a reasonable starting point for a document that will be read on screen. For email, 120–150 DPI is usually enough. For printing, stay at 200 DPI and a higher quality.
Compress, then open the result. Check the page count first. Then zoom to 100 % and read a paragraph. If the text is blurry, increase the DPI or quality and try again. If the file is still too large, reduce the number of pages before reducing the resolution further.
Settings that make sense
For a text PDF, structural mode has no quality setting because it does not re-encode the content. If it does not save enough, the problem is likely an image inside the document, not the text.
For a scanned PDF, 150 DPI at quality 0.7 is a good balance for reading and email. 100 DPI is acceptable for a receipt or a document that will only be viewed on a phone, but small print may become unreadable. 200 DPI at 0.8–0.9 is better for printing. Below 100 DPI, text starts to fall apart, especially Arabic, Chinese or small legal print.
Quality 0.7 does not mean “70 % of the original”. JPEG quality is a compression setting, not a percentage of visual quality. A lower value creates more compression artefacts. Test one page before compressing a 100-page file.
What compression can break
Text selection and search are the first casualties of image mode. If the recipient needs to copy a clause, search an invoice number or run OCR later, keep the text layer.
Links and bookmarks can disappear when pages are rebuilt as images. A table of contents that used to jump to a section may stop working.
Form fields and annotations may be flattened or lost. If the PDF still needs to be filled or signed, do not rasterise it.
Digital signatures will not validate after the pages change. If the document is already signed, compressing it can invalidate the signature even if the visible mark remains.
Accessibility suffers when text becomes an image. A screen reader has nothing to read. If the document must be accessible, use structural compression or optimise the images without rasterising the pages.
PDFs with text and large images
A report can have selectable text and a few high-resolution photos. Structural mode may save little because the photos dominate the size. Image mode will shrink it, but it will also turn the text into an image. If the text must stay searchable, compress the images themselves in an editor, then rebuild the PDF. If you only need a readable copy, image mode is acceptable; keep the original for the searchable version.
When not to compress
Do not compress a document that is already within the recipient’s limit. Every unnecessary re-encode is a chance to make it worse.
Do not compress a PDF that will be printed at high quality, especially photographs or architectural drawings. Print resolution matters more than a few megabytes.
Do not compress a signed contract before sending it to a party that will validate the signature. If you need a smaller copy for reading, keep the original as the official version and label the compressed file as a preview.
Do not compress a PDF/A archive or a document with an embedded file you need. Rebuilding the file can break the archive’s conformance.
Alternatives before lowering quality
If the file is too large because of pages you do not need, use Split PDF or Remove Pages first. Sending 20 relevant pages is better than compressing 80 pages into mush.
If the PDF contains one huge image, extract it or replace it with a smaller version in an editor. If the PDF contains several documents, split them and send only the one the recipient asked for.
If the recipient only needs to read the document, a PDF is not always the best format. For a quick preview, a JPG of the first page may be enough. For a text document, a link to a cloud file can avoid the attachment limit entirely, provided the recipient can access it.
How to check the result
Open the compressed PDF in a real reader, not only the browser tab where you compressed it. Check the page count, the first and last page, and a page in the middle.
Try selecting a sentence. If you need searchable text, this is the test. Try a link or bookmark. Try a form field if the document has one. If the document is signed, validate the signature in the reader.
Look at the file size after download, not only in the tool. Some systems report decimal megabytes while email limits use binary megabytes, so a file that looks like 4.9 MB may still be rejected at 5 MB. A compressed file that cannot be opened is worse than a large one.
Privacy
OpenToola compresses the PDF in the browser. The file is not uploaded, and there is no account or cloud copy. That is useful for contracts, medical documents and identity papers.
The browser still needs enough memory to render the pages in image mode. A very large scan on a phone may fail. Close other tabs or use a computer. Delete the compressed file after sending it if it contains sensitive information.
Tools mentioned
Frequently asked questions
Does compressing a PDF always reduce its size?
No. A well-made PDF may already be close to its minimum. Structural compression can save a little; image compression can save more but changes how the pages are stored.
Will the text still be selectable?
Structural compression keeps the text. Image compression turns pages into pictures, so text is no longer selectable or searchable.
Are my files uploaded?
No. OpenToola processes the PDF in your browser. The file is not sent to a server.