Google plans to end Chrome’s support of third-party cookies by 2022. FLoC is one of the solutions proposed to replace third-party cookies.
First, the good news.
According to Google, FLoC can provide an effective replacement signal for third-party cookies. Their tests proved FLoC to be 95% as effective as third-party cookies in terms of conversions per dollar spent.
What does FLoC mean?
It’s an acronym for Federated Learning of Cohorts.
The idea is to improve privacy by letting advertisers target groups (or “FLoCs”) of users based on common interests, rather than using an identifier for each individual user.
In basic terms, here’s how this system works.
The innovation is to train a machine learning model without a centralized repository of data.
Indeed, the model is trained on your own phone:
Your device downloads the current model, improves it by learning from data on your phone (such as browsing history, content of the websites visited), and then summarizes the changes as a small focused update.
The data stays on your phone. Only this update to the model is sent to the cloud, using encrypted communication, where it is immediately averaged with other users’ updates to improve the shared model.
Then, to preserve anonymity, these model updates are clustered into large groups of people with similar interests.
But it gets more complicated than this.
The ultimate solution hasn’t been found yet, but it’s a start.
How do you retarget an anonymous user?
The solution proposed by Google is TURTLEDOVE.
It stands for “Two Uncorrelated Requests, Then Locally-Executed Decision On Victory.”
The core innovation is to store the data that builds the retargeting lists in the user’s browser or with an independent third-party. And not on the ad network.
It’s not as easy as it sounds (and that might not have sounded easy, anyway).
Google’s engineers are working hard to find a solution for a post-cookies world.