Published on · 3 min read
Your site photos reveal your GPS position
A photo taken on a phone, a Word quotation, an exported PDF: each one carries information you never chose to send. What your files really contain, and how to clean them in ten seconds.
Take a photo with your phone, right now. It does not only contain an image: it also contains the GPS coordinates of where you are standing, accurate to a few metres, the date and time to the second, the device model, and sometimes its serial number.
This information is called metadata. It travels with the file, silently, every time you send or publish it.
What your files carry
A photo keeps the GPS position, the timestamp, the make and model of the device, the shooting settings, and often a thumbnail of the original image — before cropping. Blurring a face or cropping out a number plate is therefore not always enough: the unedited version may still be inside the file.
A Word or Excel document keeps the author’s name, the name of the last person to save it, the total editing time, and sometimes the revision history. If your quotation was built by duplicating one written for another client, that client’s name may still be in there.
A PDF records the software that produced it, the creation date, and frequently the full path of the source file on disk — which reveals your folder structure, your user name, and sometimes how your company is organised internally.
Three concrete situations
The quotation that says too much. A tradesman sends a quotation in Word. The client opens it, looks at the document properties, and discovers the name of the competitor the template was originally created for — along with the four hours spent editing it.
The published site photo. A company publishes photos of its completed work on its website. Each image carries the GPS coordinates of a client’s home. Put together, they form a list of addresses anyone can exploit.
The “anonymised” document. A report is sent out with names blacked out using the highlighter tool in a word processor, then exported to PDF. The text is still there, underneath the rectangle: select it and copy it.
The regulatory side
As soon as a photo makes a person identifiable, or locates a client’s home, you are processing personal data. The GDPR then requires you to keep only what is necessary for your purpose. Publishing a photo of completed work requires neither GPS coordinates, nor the author’s identity, nor the phone model.
Cleaning files before publication is therefore not merely a precaution: it is the data minimisation principle applied directly.
How to clean them, in practice
On Windows, the simplest route is built in: right-click the file, Properties, the Details tab, then Remove Properties and Personal Information. Windows offers to create a cleaned copy, which preserves the original. This works for photos and Office documents alike.
For Office documents, also go through File > Info > Check for Issues > Inspect Document. The inspector finds forgotten comments, tracked changes and properties.
For a genuinely anonymous PDF, never rely on visual masking: delete the text, then export. If the document will be widely distributed, export it as images before rebuilding the PDF.
For batch processing, dedicated tools quickly become essential — I ended up writing my own, because cleaning three hundred site photos one by one is not a realistic option.
The habit to build into the company
The rule fits in one sentence: every file that leaves the company is cleaned before it leaves.
That covers photos published on the website and on social media, quotations and invoices sent by email, documents uploaded to a shared space, and attachments sent to an insurer or an expert.
Ten seconds per file, and an entire category of leaks disappears. It is one of the rare security habits that costs nothing, slows nobody down, and takes effect immediately.
Try it yourself
If you want to see what your own files really contain, run this test: take the last photo published on your website, download it, and look at its properties. The answer is usually instructive — and that is generally the moment the question becomes concrete for a whole team.