Remove Metadata from Photos and Documents

How it works

1. Drop your photos, PDFs or Word documents onto the box, choose them, or paste with Ctrl+V.
2. Each file is read on your machine and the blocks that hold hidden data are found.
3. Press Clean. Those blocks are taken out and the rest of the file is left as it was.
4. Save each file from its tile. The tile also names anything deliberately left in.

About this tool

Files carry more than what you can see in them. A photograph records the coordinates of where it was taken, the camera and the exact time. A PDF or a Word document records who wrote it, who last saved it, which company they work for and which software they used. All of it travels with the file every time you send it. This removes those blocks and leaves everything else exactly as it was: photographs are not re-compressed, and a PDF keeps real, selectable text rather than being flattened into pictures. Where a name is part of the content rather than the properties, on a comment or a tracked change, it is reported instead of being quietly edited out. It runs in your browser, so nothing is uploaded.

What is actually hidden in a file?

A photograph is not only the picture. Cameras and phones write a block of data into the same file, and it usually holds the date and time to the second, the make and model of the camera, the lens and exposure settings, and, if location was switched on, the coordinates of the spot you were standing on. Some cameras add a serial number, and editing software adds its own name and sometimes the registered owner's.

None of it is visible when you look at the photo, and all of it travels with the file. Sending a picture to a marketplace listing, a forum or a stranger sends the lot. Most social networks strip it on upload, which is why this surprises people the first time they send a photo directly instead.

Documents keep a different set, and often a more awkward one. A PDF records its title, its author, the software that produced it and the dates it was made and last changed. A Word file adds the person who last saved it, the company they work for, the template it came from and how many minutes it has been open. A document forwarded outside an organisation carries all of that with it.

Does this change the file?

No, and not in the way that phrase is usually meant. The photo is never decoded and never re-saved. This reads the file's own structure, cuts out the blocks that hold the hidden data, and copies every remaining byte across unchanged. The image data that comes out is the image data that went in, bit for bit.

That is worth being precise about, because the usual advice, convert it to another format and the metadata goes with it, does work but re-compresses the photograph on the way. You lose a little quality to solve a problem that was never about pixels. Here the file simply gets smaller by however much the hidden block weighed.

A PDF is treated the same way in spirit: the document is rewritten around the same pages rather than re-rendered, so the text in it is still text, still selectable and still searchable. That is worth saying because it is the opposite of what compressing a PDF does, which turns every page into a picture. A Word file is repacked with its properties emptied and every other part copied across untouched, so the document opens saying exactly what it said before.

What is kept, and why

Colour profiles stay. A profile tells your screen how to interpret the colours in the file, and a wide-gamut photograph with its profile removed renders flat or over-saturated. It is not personal information and removing it would be a visible change to your picture, which is the one thing this tool promises not to make. Tools that strip the profile along with everything else are quietly degrading the photo to tidy up a field nobody was worried about.

One tag is kept for the same reason, and it is the one most tools get wrong. A photo taken with a phone held sideways is stored the way the sensor read it, with a tag saying which way up it goes; every viewer reads that tag and rotates the picture on screen. Strip it out with everything else and the photo suddenly appears rotated ninety degrees. So the orientation is read first and put back on its own, in a block of twenty-six bytes. Where you were standing, what took the photo and when are still gone.

Names this does not remove, and tells you about

Some names are part of what a document says rather than what it records about itself. The person credited on a PDF comment, and the author on a tracked change in Word, are both shown to anyone who opens the file. Removing them would edit the document rather than clean it, so they are found and named on the tile instead.

This matters more than it sounds. Somebody who reads half the result and sees "removed author and company names" will believe their name is gone from a document where it is still sitting in the margin. So the tile says both things on the same line, and if you see still carries names on comments, that is the part you have to deal with in Word or your PDF reader.

What this does not clean yet

Photographs in JPG, PNG and WebP, PDFs, and Word .docx files. HEIC, and other formats a browser can read but not take apart safely, are refused rather than half-done, if you have one, convert it to JPG first and clean the result. The older .doc format is not read anywhere on this site. Nor are spreadsheets or presentations, which keep properties of exactly the same kind.

There is one more honest caveat, and it is about the rest of this site rather than this page. Converting or resizing a photo here does remove its metadata as a side effect, because those tools redraw the picture, but two paths do not: a JPG placed into a PDF or a Word document is copied in whole, hidden data and all, and compressing a photo that is already as small as it can get hands your original file straight back. If what you want is the metadata gone, use this page rather than relying on a side effect of another one. The longer explanation is here.

The same is true of documents, in a smaller way. Every PDF this site writes is stamped by the library that writes it, with its name and the time the file was made; every Word file gets a creation timestamp. Cleaning a file afterwards removes those too, which is worth knowing if you have just converted something here and intend to send it on.