Compress a PDF without turning the text into a picture

How it works

1. Upload the PDF and leave the defaults alone. One file at a time. Balanced at 150 dpi is the setting that rescues most email attachments without anyone noticing the difference, so start there and only reach for the others if it is not enough.
2. Read the numbers it gives back. The result states what the file was, what it is now, and how many pictures were re-saved. A document that is mostly text barely moves, and the notes say so rather than implying the tool did something.
3. Look at the heaviest page before you keep it. Open the page with the largest photograph at full size. If the softening bothers you, run the original again at a gentler level or a higher dpi — nothing is lost, because the file you uploaded is still on your disk.

What the options do

The form above lists the options this tool is running today, straight from the server. This is what they are for.

How hard to squeeze

Sets the quality pictures are re-saved at, and nothing else: text and line artwork are untouched at every level. Gently is for a document whose photographs are the point, balanced is the one to try first, and hard makes the smallest file with softening you can see if you look for it.

Reduce pictures above (dots per inch)

The resolution ceiling. A picture is measured by how large it is drawn on the page rather than by its pixel count, so a 4000-pixel photograph placed four inches wide counts as 1000 dpi and gets scaled down, while the same photograph filling a poster does not. 150 suits screens and office printers, 300 is the number to keep if the file is going to a print shop, and much below 100 the pictures start to look like pictures of pictures.

Turn pictures grey

Removes the colour from photographs and scans. On a colour scan of a black-and-white original this is the single largest saving available, often larger than anything the quality setting can do. It is off by default because it changes how the document looks, and it cannot be undone afterwards — a grey scan does not remember what colour it was.

Remove the document details

Clears the title, author, subject, keywords and creator fields. Worth ticking before a document leaves the building, because those fields routinely carry the name of whoever wrote the template it started life as. It does not touch a word of the text, and it is not redaction.

What this tool does not do

• The saving comes from the pictures inside the file. A document that is all text has nothing much to give up, so expect a few per cent from the tidier structure and nothing more.
• It never rasterises the pages. Turning a document into pictures of itself would shrink almost anything and would also destroy the text — no selecting, no searching, no screen reader — so that trade is not offered here.
• If compressing would produce a larger file than the one you uploaded, which happens with already-optimised documents, the original is returned untouched and the notes say why.
• It does not remove pages, deduplicate repeated images or subset the embedded fonts, and it does not touch text or line artwork at any setting.
• The detail thrown out of a re-saved picture is gone from the result. Keep your original if the photographs might matter later.
• It cannot open a password-protected file. Unlock it first, with the password you already have.

What happens to your file

The file is uploaded because Compress PDF runs on our server rather than in the browser tab. It is kept for at most 2 hours and the result you download for 24 hours, then a scheduled job deletes both — a job has to be able to read the file while it runs, and you have to be able to retry a failed conversion without re-uploading. Neither needs longer.

Nothing about the contents is logged, and no one reads your documents. The retention page lists the same numbers the code enforces, and the privacy policy says who can reach the storage they sit in.

Questions

How much smaller will my file actually get?

It depends entirely on what is inside it. A 300 dpi scan typically loses half to three quarters of its size at the default settings; a text report with two logos in it might lose five per cent. The result states the before and after sizes, so you can decide whether a harder setting is worth trying.

Will the text still be selectable and searchable afterwards?

Yes. Text and vector artwork are copied across as they are, and only the raster images are re-encoded. That is the main reason to use this rather than printing the file back out to a PDF, which flattens everything into pictures of pages.

My file came back exactly the same size.

Then it was already as small as this tool can make it, and rather than hand you a larger file it returned your original bytes. That happens with documents that are all text and with files another optimiser has already been through.

Which setting should I use for a scan I have to email?

Balanced at 150 dpi first. If it is still too large, tick ‘turn pictures grey’ before you reach for the hard setting: on a colour scan of a black-and-white original the greyscale saving is bigger and the visible damage is smaller.

Does compressing break a password or a signature?

It cannot open an encrypted file at all, so unlock it first. A digital signature will not survive, because signing covers the exact bytes of the file it was applied to and this writes a new one.

It is still over my upload limit. What now?

Split by size cuts the document into parts that each stay under a ceiling you set, which is what most portals actually want. Compress first, then split, so each part carries the smaller pictures.