Compress a PDF: A Beginner's Guide
Quick answer
Open Compress a PDF, choose the file, leave JPEG quality at 70 and resolution at 110 dpi, and press Run. The tool rebuilds every page as a JPEG at that resolution and quality, then assembles a new PDF at the original page size. The new file's size is printed next to the download. If the result is too small to read, raise the quality or the resolution and run it again — you are choosing a point on one trade-off, not applying one fixed setting.
What this article covers
This is the first article in the compress-pdf cluster. It explains what the tool does to a file, the three settings it exposes, the order to change them in, and what to look at after the download. It also says plainly when compressing a PDF is the wrong move, because the tool only helps one kind of document well. The other four articles in the cluster go deeper: the FAQ and pre-flight checklist, the settings for specific jobs, business workflows, and the failure modes with their fixes.
What it actually does
A PDF page can hold text as text, or as a picture of text. A scanned document is the second kind: the page is a photograph, and its bytes are mostly image data. This tool re-encodes that image. Resolution decides how many pixels each page is drawn with, and JPEG quality decides how hard those pixels are compressed. A lower number in either field produces a smaller file and a softer page.
Nothing about the layout changes. Page count, page size and the page tree are preserved, so a document that was 12 pages of A4 is still 12 pages of A4. What is gone is the text layer: the output is an image inside a PDF wrapper, so you cannot select or search the words any more. That is not a side effect you can switch off — it is the mechanism. The page is redrawn, and a redrawn page carries pixels, not characters.
There is no upload. The browser reads the file you chose, does the work on this page with JavaScript, and hands you the bytes back. The original file is untouched.
Step by step
- Open the tool page and choose your PDF from the file picker. One file at a time.
- Set JPEG quality. Start at 70. This is the strongest lever on size and the first thing to reduce if the file is still too big.
- Set resolution. 110 dpi is a sensible default for a document that will be read on screen. For print, go higher.
- Leave Grayscale at no unless the document really is black and white. Turning it on cuts size further and is honest for a scanned contract, but it ruins a brochure.
- Press Run and wait. The page count matters: up to 50 pages is the limit for this tool, because every page has to be drawn.
- Read the size printed next to the result before you download. If it is not smaller than the input, the input was not the kind of PDF this tool helps.
What to check after download
Open the new file and look at three things, in this order.
- Small print. The smallest text on the busiest page is where quality loss shows first. Zoom to 100 per cent and read a footnote. If it is fuzzy, quality is too low.
- Page geometry. Check that the page count and orientation survived. A rotated scan should still be rotated the way you expect.
- File size. Compare it with the original. A modest reduction on a scanned document is normal; a dramatic one usually means quality went too far.
Then open it in a second viewer if you can. A file that renders in one application and not another is a sign that the input was unusual rather than that compression failed, and it is worth knowing before you send it to a client.
Common beginner mistakes
The most common mistake is running this on a born-digital PDF — one exported from a word processor, with real selectable text. Those files are already compact. Re-encoding their pages as pictures can make them larger and always makes them unsearchable. If you can select a sentence in the original, this is usually the wrong tool.
The second mistake is turning quality down until the file is small, without reading the result. Quality 20 on a page of 9-point type produces a file nobody can read. Compress to a target you have checked, not to the smallest number the slider allows.
The third is forgetting the page limit. A 120-page scan is refused rather than processed, and the message says so. Split the document first, then compress the parts.
When this tool is the right choice
Use it when the page is already a picture: a scan, a photographed document, a fax, a screenshot printed to PDF. Use it when the destination has a hard size limit — an email attachment cap, a portal upload limit, a form that rejects anything over a few megabytes. Use it when the document only needs to be read, not searched or copied.
Do not use it when the recipient needs to select text, when the file must remain accessible to a screen reader, or when the PDF is a form with fields you still want to fill. Those need a different answer, and this page does not pretend to be one.
Related tools
Compression is often one step in a longer chain. If you are assembling a set of documents, Merge PDF Files joins several PDFs into one and keeps their text, links and form fields, because it copies pages instead of redrawing them. If the destination wants pictures rather than a PDF, PDF Converter renders the pages you choose as one JPEG per page inside a ZIP. Both are honest neighbours of this tool and neither replaces it.
Use this tool
Go to Compress a PDF, pick the file, and start with quality 70 and 110 dpi. Change one setting at a time so you know which one produced the result you are looking at.
Next step
Read the FAQ and checklist for this tool before you send anything important. It lists the checks that catch the two failures beginners actually hit: a file that is smaller but unreadable, and a file that is smaller but no longer the document you thought it was.
Continue Exploring
Browse more release notes and the tools they reference.