How does a web beacon(web bug) work?

When an HTML page is downloaded the browser parses the page and looks for additional resources needed to display the page, such as images. For each image it finds the browser makes another request to a server in the background. When servers receive requests, they usually log the request to monitor load on the server, and record information about who sent the request and where it came from. A web beacon is a tiny invisible graphic that generates a request to the tracking firm's server. They record the request in their logs and then analyze their logs to see who went where and did what and when.

When returning the image from their servers to the browser, they can also send down information to be added to a cookie. There are third-party cookies that can be tracked across domains. If you come back to the site, and the beacon request is made again, that cookie will also be sent up in the request to the server and the tracking firm will have more information about you.

Think about this. Even though you are visiting myfavoritesite.com the web beacon image is being requested from trackers.com. The cookie they create is assigned/locked to their domain, trackers.com. But if you then surf over to myotherfavoritesite.com, and they too are sending web beacons to trackers.com, the cookie will essentially be shared between the two sites. There are more considerations here, but that is the basic premise.