Skip to main content
BrowserStay

Command Palette

Search for a command to run...

PDF

Is it safe to upload a PDF to an online converter?

The honest answer is: it depends on what the document is, and on things the site does not tell you. Both halves are checkable.

Compress a PDF without uploading

Every week a document goes to a site called something like “free-pdf-tools”, gets converted, and comes back. The question people actually ask is not whether HTTPS is switched on. It is: who has a copy now, and for how long?

What the server does with it

A conventional converter works in three steps. Your browser sends the file over an encrypted connection to a machine the service rents. That machine writes it to disk, processes it with a conversion library, and sends the result back. Then there is a fourth step nobody advertises: the file sits somewhere until something deletes it.

Every stage is a design decision the service made, and the only one you can verify from outside is the first. Encryption in transit protects the file from the network between you and them. It says nothing about the disk it lands on, the backup that copies it an hour later, the logging system that recorded the filename, or whether the retention policy on the marketing page matches the cron job that cleans the bucket.

None of this makes such services villains. It makes them a place where the answer depends on their infrastructure, their staff, and their future — including who buys them next year. A policy can change on a Tuesday; a document cannot be un-copied.

The three questions that decide it

  • What is in the file? A published brochure, a blank form, a press release: upload it, nothing happens. A signed contract, a medical letter, a passport scan, a payslip, a property deed: the file is the liability, and every copy is a new one.
  • What does the site say it keeps? Look for a retention sentence, not a privacy badge. “Files are deleted after one hour” is a claim you cannot audit. “We never receive the file” is a claim you can test in two minutes, because the network tab will show nothing leaving.
  • How much do you trust this particular site? A tool with a name, an address, and a changelog is a different bet from a site whose footer is a copyright year and a Gmail address.

What a browser-side tool changes

If the work happens in the tab, there is no upload to trust. The file is read from your disk into memory, transformed by code that shipped as part of the page, and written back out as a download. Nothing crosses the network except the page itself.

That is not a guarantee of good intentions either — a page can still load a script that phones home. Which is why the check matters more than the promise: open DevTools, watch the Network tab, drop a file in. If the only requests are scripts, fonts and images, the file stayed put. Run it on us too; that is the point of teaching the test rather than asking you to believe a badge.

When uploading is genuinely fine

There are real cases where the server is the better answer: a 900 MB scanned archive on a phone with 2 GB of free memory, a format only a desktop engine can decode, a file that is already public. The mistake is not using a server. The mistake is making it the default, with the one document that would hurt most if it leaked, because the site had a nicer logo and ranked first.

A short checklist

  1. Decide what the document is worth before you pick a tool.
  2. Prefer a tool that does not need the file. No upload, no retention question.
  3. If you must upload, read the retention line, and prefer a service that names an operator.
  4. Strip metadata before sending anything — author, software and timestamps travel with the file.
  5. When in doubt, do the network test on the site you were about to trust.