Compress PDF

A PDF is a container. It holds the text of each page, the fonts that draw that text, vector shapes such as lines and charts, and pictures. In a typical office document the text and fonts are a small part of the total. Pictures are what make a file heavy: a single page scanned in color, or a phone photo of a receipt, can weigh more than a hundred pages of plain text.

That is why every mode above works mainly on images. Each picture is decoded, scaled down when its resolution is higher than the level you chose, and saved again as a JPEG at the chosen quality. Pages that contain only text barely change, which is also why a text-only PDF cannot get much smaller no matter which mode you pick.

Scanned documents deserve a special mention. A scanner or phone app saves each page as one big picture, so even a short letter can be heavy. Recommended mode handles these well, and grayscale helps further when the paper was black and white to begin with. A scan with no text layer stays a picture after compressing; if you also want to search or copy its words, run it through text recognition before or after.

Resolution is measured in dots per inch of the printed page, so it tells you how much detail each picture keeps relative to its size on paper. When an image has more dots per inch than you allow, it is resampled down to your limit. When it already has fewer, it is left at its own resolution, because adding pixels would only make the file bigger without adding detail.

Image quality is the second dial. It controls how strongly the JPEG encoder simplifies each picture. Lower values save more space but bring blocky patches and halos around letters in scanned text. If a scan becomes hard to read, raise the quality first and lower the resolution instead; text usually survives a lower dpi better than heavy JPEG compression.

Turning on Convert to grayscale removes color from every page. For scans of black-and-white paperwork this saves a lot of space with no real loss, because a gray image stores one channel instead of three. If you only need to drop color, the dedicated PDF to Grayscale tool also offers a pure black-and-white mode.

A practical way to choose: decide where the file will be read. A document that is only opened on screens can go to 100 or even 72 dpi. One that someone may print at the office should stay near 150 dpi. Anything with fine technical drawings, signatures that must stay legible, or small footnotes in a scan is safer at 200 dpi or above, even if the file ends up a little larger.

Email is the most common reason to compress a PDF. Gmail and Outlook both publish attachment limits, shown above with links to their own help pages, and the limit applies to the whole message, not to each file. Because attachments are encoded for sending, a file that is just under the limit on disk can still be refused, so leave some room.

Upload forms are stricter. University applications, job portals, visa systems and tax offices often accept only files below a stated size, and they say so in small print beside the upload field. That number is what the Target size mode is built for. If a portal also asks for a minimum size, which happens with some government systems, Increase PDF Size does the opposite job.

When one PDF is simply too long to fit any limit, compressing it harder is not the only option. Cutting it into parts with Split PDF, for example by a maximum size per part, keeps every page at full quality and sends the document in several messages.

Besides pictures, a PDF can carry material nobody sees. Subset fonts keeps only the characters a document actually uses from each embedded font, which can make a real difference in files that embed several complete font families. The text looks exactly the same.

Remove metadata deletes the author name, the software used, edit history and XMP data. It saves a little space and also stops personal details from traveling with the file. Flatten form fields turns filled-in fields into ordinary page content, so the values can no longer be changed; use it only on a final copy. Fast web view reorders the file so a browser can show the first page before the rest has downloaded, which helps when the PDF is hosted on a website.

Compressing several files at once works the same way: every PDF in the batch gets the same mode and settings, and each one comes back as its own file. If one document needs a strict target size and the others only need the default, put it in a separate batch so the target applies to that file alone.

Keep the original file. Compression cannot be undone, and a smaller copy made for email is rarely the right one to print or archive. If the result looks too soft, run the original again with a higher resolution instead of compressing the compressed copy.