-
Notifications
You must be signed in to change notification settings - Fork 215
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
How can advertisers learn about interest groups? #33
Comments
Yes, as long as the "URL information" is made aggregatable. Certainly you should be able to use the aggregated reporting API to record your interest group and the domain of the publisher page, and then get a histogram of the top domains on which that interest group showed ads. But the counts in the histogram would include some noise, and insufficiently popular domains for the interest group wouldn't appear at all. Instead of aggregating on the raw domain, you could process the URL — either in your bidding JS directly, or on your server as part of the contextual call that produces a contextual signal that becomes auction input. Then for each interest group you could get a histogram of the values you derived from the URL. That could be a coarser grouping than domain (for example, to cluster together a bunch of small sites with similar topics) or a finer grouping (for example, to break out pages on particular topics from a large heterogeneous site). |
In our W3C meeting yesterday(June 2nd), we explained that advertisers could learn about the members of a FLoC by analyzing the behavior(URLs) observed under a given FloC as part of ad requests. Through this learning, advertisers can reach an audience by intelligently choosing which set of FloC identifiers to target.
Is there a way to achieve analogous behavior under turtledove such that advertisers can learn about interest groups? Can the aggregate reporting API be used to communicate aggregate URL information to be further analyzed by advertisers?
The text was updated successfully, but these errors were encountered: