Which method suits my PDF?
A PDF usually contains two kinds of content. Text, lines and table borders are stored in the file as descriptions and take up little space. Embedded images such as photos, logos or scanned pages, on the other hand, often make up the largest part of the file. The main difference between the three methods is whether they are allowed to change these images.
With Balanced and Smaller file, only suitable images are eligible for recompression. Very small graphics, for example, remain unchanged. If the losslessly optimized version still turns out to be the smallest, that is the version you get. None of the methods lowers the image resolution (DPI), crops margins or converts entire pages into images.
If you're unsure, start with Balanced. If the result is still too large, try Smaller file and then take a closer look at the text. Lossless optimization is worthwhile when nothing may be changed in photos, seals or graphics and you accept that the savings may be small.
Text stays text, scans stay scans
Text that you can select and copy in your PDF is not converted into an image during compression. It therefore usually remains selectable and searchable. The best way to check whether this also applies to your file is to take a quick look at the saved copy.
A scanned page is different, because it is already an image. Text recognition (OCR) is not part of this tool. A scan without a text layer therefore doesn't become searchable through compression. Because the entire page of a full-page scan is contained in this image, the entire page content is affected when Balanced or Smaller file actually recompresses the image. Whether that happens depends on whether the image is suitable for it. If a scan has been recompressed, you should check small text particularly closely.
Reducing PDF size: step by step
- Tap or click “Choose PDF” or drag the file into the box. On a smartphone, this opens your device's file picker. The tool processes exactly one PDF of up to 50 MiB, i.e. 52,428,800 bytes, per run.
- Choose the method. Balanced is the default.
- If you need to stay within a fixed limit, tick “Set target size” and choose a value (see below). Without a limit, you can skip this step.
- Start the compression. If a run takes too long, you can cancel it.
- Check how large the original and the result each are, and page through all pages in the comparison.
- Download the new file and open it separately once before you send or upload it.
If you change the method or the limit after the run, the tool discards the previous result. That way, the file shown always matches the currently selected settings. Then simply start the compression again.
Setting a maximum size
The available limits are 500 KB, 1 MB, 2 MB, 5 MB and 10 MB. You can also enter your own value in KB, MB or MiB. The tool calculates in decimal units: 2 MB is 2,000,000 bytes and 500 KB is 500,000 bytes. 2 MiB, by contrast, is 2,097,152 bytes. The target counts as reached if the result has no more bytes than your limit.
The target size is therefore an upper limit: the result should be at most that large, not exactly that large. It isn't reached in every case. If your PDF is already below it, the image quality isn't touched. The rewritten structure may still make the file somewhat smaller. If it doesn't work, you still get a complete, valid PDF and can see that the limit was missed. To find out how to proceed from there, see Reduce a PDF to fit a size limit. If you're not sure which unit a requirement refers to, the calculator helps you tell KB, MB and MiB apart.
Comparing the original and the result properly
After the run, you see the file size of the original and that of the result, each also in bytes. You also see the savings: how many bytes were saved and what percentage that represents. If you set a limit, the tool also shows you whether it was reached. However, the figures only tell you whether the file is small enough. Whether it is still easy to read only becomes clear when you compare the pages.
In the comparison, the original and the result are shown side by side. Use the arrows to page forward and back. Don't just look at the first page, because photos, scans and appendices are often further back. Pay particular attention to these areas:
- small text, footnotes and fine print
- stamps, seals and scanned signatures
- tables, charts and fine lines
- photos in which details must remain recognizable, such as scans of ID cards or certificates
Afterwards, open the downloaded file again in your usual PDF program. Select a word that you know exists as text, and search for it. Also click any links that need to work, and check filled-in form fields. A page that looks the same doesn't tell you whether all of your file's functions have been preserved.
With this test file, lossless optimization saved hardly anything, while the other two methods saved a great deal. The difference here comes from image compression, because only Balanced and Smaller file are allowed to recompress suitable images such as the included test image. For a PDF with different content, such as mostly text without suitable images, the values can look completely different. How much the optimization of structure and data streams achieves there depends on the particular file.
When the file stays unchanged
Some PDFs can hardly be reduced, or not at all. Often the file is already stored efficiently, its images are already heavily compressed, or they are very small graphics that are not recompressed. If a PDF contains mainly text and no suitable images, only the optimization of structure and data streams can make a difference. How much that achieves depends on the particular file. The tool discards newly optimized versions that are the same size as or larger than the original. If no valid new version is smaller, it reports that the file remains unchanged and shows your original as the result, at its unchanged size. You can then download it as usual.
These next steps make sense:
- Try a different method, for example Balanced after Lossless optimization, or Smaller file after Balanced. Always start from the original and not from an already reduced result, because repeated JPEG compression degrades images step by step.
- You can remove pages from the PDF that nobody really needs, such as blank pages or duplicates.
- If the recipient accepts multiple files, you can split the PDF into suitable parts.
- Some portals limit not only the size of a file but also the number of files, the page count or the total size. You can check these other upload limits.
Converting a PDF into JPG images, on the other hand, is not a good substitute for compression. The recipient usually expects a PDF, and selectable text would be lost.
What happens to my file?
Compression runs in your browser's memory. Your PDF is not uploaded to a server for this, and you don't need an account either. The tool works with a copy, so the file on your device is not overwritten. You save the result as a new file using your browser's download function. No watermark is added.
Local doesn't mean offline, though. Your browser loads the page and the program components for processing from the internet. You therefore shouldn't assume that everything will also start without a connection.
With large files, how long processing takes also depends on how much memory your device currently has available. If you are working on documents for an employer, school or university and there are rules about which programs may be used to process confidential documents, those rules take precedence.
Frequently asked questions about reducing PDF size
Signed and protected files
The tool rejects digitally signed or certified PDFs, as well as encrypted and password-protected files. A digital signature applies to exactly the version that was signed. A rewritten file would no longer be the same. The tool does not remove passwords. If you are permitted to, use an unsigned and unprotected version, such as the file before it was signed, or ask the sender for one. Empty signature fields that have not yet been signed, on the other hand, are not an obstacle.
Can I compress several PDFs at once?
No, the tool processes exactly one PDF per run. If you select several files at the same time, the selection is rejected. Process the files one after another. For other tasks, such as merging, you'll find more PDF tools.
How do I start with a new file?
Reset the tool. The default settings then apply again, with Balanced and no limit, and you can select a new PDF. Download the previous result beforehand if you still need it.