You have a contract to sign, a passport scan to shrink, a bank statement to convert. You search for a free tool, drag the file in, and get your result back a few seconds later. The question nobody asks at that moment is the one that matters: where did the file just go?
Two completely different architectures
Online tools split into two kinds, and they are not distinguishable from the outside. The interface looks identical either way.
Server-side: your file is uploaded
The page sends your file over the network to a machine you know nothing about. It is processed there and the result is sent back. From the moment the upload completes, the file exists on hardware you do not control, run by a company you have not checked, in a country you did not choose.
What happens to it next depends entirely on the operator’s honesty:
- It sits in a temporary directory until some cleanup job runs — hours, or never.
- It is backed up, because production servers usually are.
- It passes through a CDN or proxy that logs and caches.
- It is retained deliberately — for “quality improvement”, or as training data.
A privacy policy saying “files are deleted after one hour” is a promise, not a mechanism. You cannot verify it, and it does not survive a breach, an acquisition, or a subpoena.
Client-side: the file never leaves
The page loads a program into your browser, and that program does the work on your own machine. The file is read from disk by the page itself, processed in memory, and written back out. Nothing is transmitted, so there is nothing to retain, log, cache or leak.
This is not a policy. It is an absence of the capability, and that is a much stronger guarantee than any promise.
Why doesn’t everyone do it this way?
Partly history — browsers only became capable enough fairly recently, and a lot of these sites are older than that. Partly economics: if the file is on your server you can build a business around it, meter it, require an account for it.
And partly because some things genuinely cannot be done in a browser. A live currency rate has to come from somewhere. A video platform’s thumbnail lives on their servers, not yours. Those are honest exceptions. Converting a PDF to images is not one of them.
How to tell which you are using
You do not have to take anyone’s word for it, including ours. Open your browser’s developer tools — F12, or Cmd+Option+I on a Mac — and click the Network tab. Then use the tool.
- Watch for a large upload. A request carrying your file’s worth of data is unmistakable: sort by size and it will be at the top. If your 8MB PDF produced an 8MB request, it was uploaded.
- Try it offline. Load the page, disconnect from the internet, then use the tool. If it still works, the work is happening on your machine — it cannot be anywhere else.
The offline test is the definitive one. It cannot be faked, and it takes about fifteen seconds.
Other signals worth noticing
- A progress bar that scales with your connection rather than your file size means uploading.
- “We’ll email you the file” means it is on a server, and now so is your address.
- An account requirement to download your own result is a business model, not a technical need.
- A file size limit usually means server-side — in-browser processing is limited by your own memory, not their disk quota.
Where this actually matters
For a holiday photo, honestly, it does not. The reason to care is what people actually put through these tools:
- Signed contracts and NDAs
- Passport and identity scans, for visa and job applications
- Bank statements, for mortgage and rental applications
- Medical letters and test results
- Unreleased designs, screenshots of internal systems, draft filings
If your employer has a policy about where company documents may be sent, an online converter is squarely covered by it — and it is the kind of thing that is only noticed afterwards.
Photographs deserve a specific mention, because they carry more than they appear to. A photo from a phone usually records the exact coordinates where it was taken. You can see what yours contain and strip it — and note that a tool asking you to upload a photo in order to remove its location data has already done the thing you were trying to prevent.
How this site is built
185 of the 187 tools here run entirely in your browser. The remaining 2 need something a browser genuinely cannot do alone, and those pages say so at the top rather than in a policy.
Please do not take that on trust. Open the Network tab, or pull the cable. That is the point of building it this way.