-
Myles_Younger It depends on what else the publisher passes into the ad request beyond the obvious TLD that would have to be there. I agree that generally clean room matching obliterates any privacy promise. Which is why I would consider DNS or GDPR opt-out to opt one out of a clean room.Permalink On twitter.com
Mood +6 🙂
-
Myles_Younger If you are asking 'does clean room style matching recreate all the privacy problems of third party cookies but with slightly more limitation in terms of who gets access to user data' the answer is yes...Permalink On twitter.com
❤️ 5 Favorites
Mood 0
-
Myles_Younger UID2 and LiveRamp as a specifications provides optional methods to reduce data leakage with double encryption... in theory. But your target advertiser will still get all the advantages of knowing that user in your context and the capacity to join on that with other contexts.Permalink On twitter.com
❤️ 2 Favorites
Mood +3 🙂
-
Myles_Younger Unclear to me what level of encryption PAIR requires. I would consider 'hashed email' straight up user data leakage. Hashing isn't sufficient security client-side and most (but not all) hash methods aren't sufficient anywhere else either...Permalink On twitter.com
❤️ 3 Favorites
Mood 0
-
Myles_Younger None of these systems are anonymizing users sufficiently to meet any legal requirements, so all opt outs apply. But... that's sort of the point of these systems. To me a big question is more how do publishers isolate their matches against reuse to prevent the data leakage of 3p?On twitter.com
♻️ 1 Retweets
❤️ 5 Favorites
Mood +1 🙂
-
Myles_Younger (BTW: you wouldn't need UTMs or click trackers in the current system. Just precise timing data is enough. UTMs would do a LOT but they are *slowly* on their way out.)Permalink On twitter.com
❤️ 3 Favorites
Mood 0