Google to address ad frequency with machine learning when cookies aren’t available
The feature will launch first in Display & Video and eventually come to Google Ads.
Google will be rolling out an approach to ad frequency control that doesn’t rely on cookies in Display & Video 360 over the coming weeks. The company said Tuesday that it plans to bring it to Google Ads in the future.
The feature uses machine learning to analyze traffic patterns when third-party cookies are available and builds models to predict patterns when a cookie isn’t present. “This allows us to estimate how likely it is for users to visit different publishers who are serving the same ads through Google Ad Manager. Then, when there is no third-party cookie present, we’re able to optimize how often those ads should be shown to users,” said Rahul Srinivasan, product manager for ads privacy at Google, in the announcement.
Why we should care
Google is increasingly relying on models to inform how ads are delivered when it doesn’t have access to data it could once count on. Google is in an enviable position, however: the massive volume of data it’s still able to collect combined with significant investments in machine learning means it can make do with less.
More on the news
- The company says it aggregates user data before applying its machine learning models, so no user-level data is shared across sites and relies on publishers’ first-party data.
- The company also says the feature “respects a user’s choice to
opt outof third-party tracking.” Google and other digital advertising companies have long ignored Do Not Track.
- Chrome announced changes to the way it is handling cookies and fingerprinting in May. It will start requiring developers to specify which cookies are able to work across sites and potentially used to track users.
Opinions expressed in this article are those of the guest author and not necessarily MarTech Today. Staff authors are listed here.