At DuckDuckGo, we don't track you, ever. That's our Privacy Policy in a nutshell. To improve our product, we develop completely anonymous ways to figure out how our products are being used. We do not create unique cookies or store IP addresses, and, more generally, we architect our product and these systems so that we do not even have the ability to create a search or browsing history for any individual — it's privacy by design.
What We Collect
When YouTube ad blocking is turned on, the DuckDuckGo browser checks for a few specific conditions that can tell us when ad blocking may not be working as expected:
- YouTube showing a message indicating that ad blockers are not allowed or asking the user to disable ad blocking.
- A YouTube player error, such as "this video is unavailable" or an unusual traffic challenge.
- A long buffering pause before video playback continues.
- A video or static ad being shown despite ad blocking being on.
When any of these are detected, the browser sends a single anonymous request, at most once per category per device per day, so we can estimate how often each condition occurs when this feature is on. Each request also includes whether or not the YouTube account appears to be signed in and, if it is, whether it’s a standard or Premium account. This anonymous signal allows us to see if issues affect certain user states differently.
What We Don't Collect
These requests never include the URL, title, channel, or any other identifying information about the video being watched. They also never include your YouTube account name, search history, or browsing history. They do not include any unique identifier that could be linked back to you. The requests are fully encrypted in transit, and we do not store IP addresses associated with them.
How This Helps Improve YouTube Ad Blocking
YouTube actively works to detect and block ad blockers, and the techniques they use change over time. Without this anonymous diagnostic data, we have no way to know if ad blocking has stopped working for our users, how widespread an issue is, or whether a fix has successfully addressed the issue. Counting these events anonymously lets us spot new YouTube detection methods affecting users, prioritize fixes based on how many people are affected, and verify that our updates actually resolve issues.