The problem with lists

Most blocking tools work from a list. Somebody maintains a catalogue of known tracking domains, your browser downloads it, and anything on the list is refused. This works well and it has a structural weakness: the list is always slightly behind. A tracker that rotates its domain names, or one that only appears on a handful of sites, can operate in the gap between being deployed and being catalogued. Since the economics of tracking reward novelty, that gap is where a lot of the activity happens.

Privacy Badger takes a different route. It does not ask whether a domain is on a list. It watches what a domain does. If the same third party appears across unrelated sites you visit, and it is storing identifiers that let it recognise you on the next one, the extension treats that behaviour as the signal and starts blocking. No configuration, no subscription, and nothing to keep up to date on your side.

Behaviour as the criterion

The mechanism is simple enough to explain in a sentence and clever enough to be worth the trouble. Every time you load a page, the extension records which third party domains were contacted. A domain seen on one site is unremarkable; a domain seen on many unrelated sites, storing the same identifier each time, is a tracker by definition rather than by reputation. The blocking decision follows the observation, which means the tool adapts to behaviour instead of to a database.

Two smaller features are worth knowing. Social buttons embedded in pages are a tracking channel in their own right, because the script loads whether or not you click, so the extension neutralises them until you actually interact. And links that arrive carrying campaign parameters get those parameters stripped, which stops the referral information from following you into the destination page.

The organisation behind it

The publisher is a digital rights organisation that has spent decades arguing for user privacy in courts and legislatures, and the tool reflects those priorities rather than a business model. It is fully open source, there is no paid tier, no analytics of its own, no commercial partnerships influencing which domains get treated leniently. That independence is the reason the tool is trusted, and it is also the reason it is not the most aggressive blocker available.

That last point matters. Tools funded by advertising interests have an obvious conflict; tools funded by donations have a different constraint, which is that they must be defensible. When the extension is unsure, it defaults to allowing rather than breaking, and it shows you what it did so you can overrule it. For a general audience that is the right trade: a page that works with slightly less protection beats a page that fails to load.

Privacy should not be a premium feature for people who know what a tracking pixel is. A free extension from an organisation with nothing to sell is the most useful thing that can be handed to an ordinary user.

How to use it well. Install it and browse normally for a week before judging, because the learning needs material to work with and the panel is more informative after it has seen real behaviour. Leave it alongside a conventional list-based blocker rather than replacing one with the other, since they catch different things. And check the panel occasionally for sites that broke: sliding a domain back to allowed takes a click, and that manual correction is the entire maintenance burden this tool imposes.