What is a tracking pixel?
How the invisible image in a marketing email reports that you opened it — and what it can and cannot see.
A tracking pixel — also called a spy pixel, open pixel or web beacon — is a tiny image embedded in an email whose real job is not to be seen. When your mail client downloads it, that download tells the sender the message was opened.
The mechanics
An HTML email is a web page. Like any web page, it can reference images that live on a remote server:
<img src="https://track.example.com/o/a8f3e1c9b2.gif"
width="1" height="1" style="display:none" alt="">
Three details make this an open-tracking pixel rather than an ordinary image:
- It is remote. The image is fetched from the sender’s server every time it is displayed, rather than being attached to the message.
- It is invisible. It is typically 1×1 pixel, transparent, or hidden with CSS, so you never notice it.
- It is unique to you. The path or query string contains an identifier (
a8f3e1c9b2above) that maps back to your specific email address and this specific campaign.
When your mail app renders the message, it issues an HTTP request for that image. The server logs the request — and because the identifier is unique, it knows who opened which email.
What it can see
A single pixel request reveals more than a yes/no “opened”:
- That you opened the message, and the date and time you did.
- Your IP address, which gives an approximate location and your network/ISP.
- Your user-agent, i.e. the mail client and often the operating system.
- Repeat opens, since each render is another request — so senders can count how many times and roughly when you re-read a message.
Rewritten links do the same thing for clicks. Instead of pointing straight at the destination, a “click” link routes through the sender’s server first, which records the click before forwarding you on. Our Link inspector shows that chain hop by hop.
What it cannot see
A tracking pixel is not a microphone or a keylogger. It cannot read the rest of your inbox, see what you type, or access files. Its power comes entirely from the fact that loading a remote resource is itself a signal. Remove the load — by blocking remote images — and the signal disappears (with important caveats covered in what image blocking really does).
Why it matters legally
Because inserting and reading a pixel means writing to and reading from your device (the request, the stored image, the cache), European regulators treat it like any other access to terminal equipment under Article 5(3) of the ePrivacy Directive. In France, that is Article 82 of the Loi Informatique et Libertés. The consequence — explored in the CNIL’s 2026 recommendation — is that most tracking pixels require your prior consent, with a few narrow exemptions.
Related guides
- Law
Exempt vs. consent-based pixels
Not every pixel needs consent. Where the line sits between security/deliverability measures and marketing analytics.
- Technical
What image blocking does (and doesn't) do
Why blocking remote images stops many pixels, why proxies are only a partial fix, and what still leaks.