All tools run in your browser — your files never leave your device.
All tools154

PDF

ZIP and PDF: what actually compresses, and what does not

Zipping a large PDF is the obvious move and it almost never works. The reason is worth understanding, because it applies to most modern files.

The short answer. ZIP compresses by finding repetition. A PDF is already compressed internally — its images are JPEG or Flate encoded and its text streams are Flate encoded — and compressed data has no repetition left to find. Zipping a typical PDF saves between 0% and 5%. ZIP is genuinely useful for bundling many files, for text and CSV, and for encryption in transit.

Why it does not work

Lossless compression works by describing repetition more briefly. If a byte sequence occurs a hundred times, store it once and reference it.

Compression is therefore the act of removing repetition. Once removed, it is gone — a compressed file looks close to random, and random data has nothing left to describe more briefly. This is not a limitation of ZIP; it is a mathematical property, and no compressor beats it.

Why it does not work
File typeAlready compressed?ZIP saves
PDFYes, internally0–5%
JPEG, PNG, WebPYes0–3%
MP4, MP3Yes0–2%
DOCX, XLSX, PPTXYes — they are ZIPs already0–2%
Plain text, CSV, JSONNo60–90%
BMP, WAV, uncompressed TIFFNo50–80%
SQL dumps, log filesNo80–95%

The DOCX row surprises people. A .docx file is a ZIP archive containing XML — renaming one to .zip and opening it works. Zipping a Word document is zipping a zip.

What ZIP is genuinely for

Three jobs, all of which it does well and none of which is making a PDF smaller.

  • Bundling. Forty files become one attachment, with the folder structure preserved. This is the main reason to use it.
  • Compressing uncompressed formats. Text, CSV, logs and SQL dumps shrink enormously — a 100 MB log file routinely becomes 5 MB.
  • Encryption in transit. ZIP AES-256 is real encryption, and it is a reasonable way to protect an attachment where the recipient cannot use anything better.
  • Getting past file-type filters, which block .exe and sometimes .pdf but usually allow .zip. Worth knowing, though it is also why some mail systems now block ZIPs.

Note the encryption case has a real limit: it protects the file until someone extracts it. After that there is an unprotected PDF on their disk. The comparison with PDF encryption is in PDF security: what actually works.

What to do instead

If the goal is a smaller PDF, compress the PDF rather than wrapping it.

The size is almost always the images inside it. Re-encoding them at 150 DPI and quality 0.75 typically removes 60–80% of a scanned document — two orders of magnitude more than ZIP will give you, and the reasoning is in how to compress a PDF without losing quality.

For a folder of photographs, the same applies: compress the images with the Image Compressor and then zip them if you need one file. Zipping first and hoping is the step that wastes time.

Does zipping lose quality?

No, and this is worth stating clearly because the question is asked constantly.

ZIP is lossless. Every byte that goes in comes out identical — that is the entire specification. A photograph extracted from a ZIP is bit-for-bit the file that went in.

The confusion comes from conflating ZIP with the lossy compression inside JPEG or MP4. Those discard data permanently. ZIP never does, which is also precisely why it cannot make an already-compressed file much smaller: it is not allowed to throw anything away.

Compression ratio, and why 100 MB became 99 MB

Compression ratio is original size divided by compressed size. A 100 MB file becoming 25 MB is 4:1, or 75% saved.

Ratios vary enormously by content, which is why a single expected number does not exist. Text reaches 10:1. A PDF reaches about 1.02:1. Anything already compressed sits near 1:1, and occasionally slightly below — a ZIP of a JPEG can be very marginally larger than the JPEG, because the archive adds headers and there is nothing to reclaim.

So a compression tool reporting "1% saved" is not broken. It is reporting the truth about a file that had nothing left to give.

ZIP problems worth recognising

  • "The archive is corrupt." Usually a truncated download. Check the size against the source before trying repair.
  • Non-English filenames become mojibake. ZIP has a long history of encoding ambiguity; extracting with a tool that supports UTF-8 filenames usually fixes it.
  • Path too long on Windows. Deep folder structures inside an archive can exceed the path limit. Extract to a short path such as C:\temp.
  • The password does not work. Older ZIP encryption and AES are different schemes; some tools create archives others cannot open. 7-Zip reads essentially all of them.
  • Empty folders vanish. Some tools skip them, which matters if the structure is the point.

And one thing that is not a problem: you cannot convert a ZIP to a PDF. A ZIP is a container that might hold anything. What people mean is converting its contents — extract first, then convert whatever is inside.

Frequently asked questions

Does zipping a PDF make it smaller?

Barely — typically 0-5%. A PDF is already compressed internally, and compressed data has no repetition left for ZIP to find. To actually shrink a PDF, re-encode the images inside it, which usually removes 60-80% of a scanned document.

Does zipping reduce quality?

No. ZIP is lossless by specification — every byte that goes in comes out identical. The confusion comes from lossy compression inside JPEG or MP4, which does discard data. ZIP never does, which is exactly why it cannot shrink an already-compressed file.

What compresses well in a ZIP?

Anything not already compressed: plain text, CSV, JSON, logs and SQL dumps reach 60-95% savings. Already-compressed formats — PDF, JPEG, MP4, and Office files, which are themselves ZIP archives — save almost nothing.

Why is my ZIP the same size as the original?

Because the contents were already compressed. This is expected rather than a fault. A ZIP of a JPEG can even be very slightly larger, since the archive adds headers and there is nothing to reclaim.

Can I convert a ZIP to a PDF?

No — a ZIP is a container that might hold anything, and there is nothing to convert. Extract it first, then convert whatever is inside. If it holds several PDFs, merge them.

Is a password-protected ZIP secure?

ZIP AES-256 is genuine encryption and protects the file in transit. Its limit is that protection ends at extraction — after that there is an unprotected file on their disk. Note also that older ZIP encryption is weak and different tools implement different schemes.

Stop reading, start doing

Every tool in this guide is free.

154 browser-based utilities. No account, no upload, and no file size limit — your files are processed on your own device and never sent anywhere.

Browse all 154 tools