What The FLoC? Will Google’s Federated Learning of Cohorts Replace 3rd Party Cookies This Year?

The History

Back in August 2019, Google announced their ‘Privacy Sandbox’ initiative, to develop open standards for improving web privacy for users, without hamstringing businesses. This was generally well received due to the greater demand for online privacy from users.

Then, in January 2020, they announced that they intend to phase out support for third party cookies in their Chrome browser within two years. This obviously raised concerns for businesses relying on cookies for ad serving. But Google tried to assuage these concerns.

“we are confident that with continued iteration and feedback, privacy-preserving and open-standard mechanisms like the Privacy Sandbox can sustain a healthy, ad-supported web in a way that will render third-party cookies obsolete”

Whilst also commenting on those browsers who had already blocked 3rd party cookies (Safari & Firefox), suggesting that that they were causing harm to businesses and users.

“By undermining the business model of many ad-supported websites, blunt approaches to cookies encourage the use of opaque techniques such as fingerprinting (an invasive workaround to replace cookies), which can actually reduce user privacy and control.” 

Where Are We Now?

So over a year on what has changed with Google, and how might this affect your marketing?

Change is definitely coming. With a new security policy to limit cross-site tracking, browsers using SharedArrayBuffers again now that they can mitigate the risk of private memory access (after previous failure), and the announcement in a blog recently that Google is working to put a stop to online fingerprinting. They’ve been busy!

Most recently Google has finally shared more about the alternative technology it’s developing to replace 3rd party cookies, born of its Privacy Sandbox. This technology is called Federated Learning of Cohorts or FLoC.

FLoC aims to move away from the current methods of cookie based targeting of individuals (which often store user information), and to instead use a far more anonymised and aggregated method of interest based targeting. This will involve users being grouped into large interest and behaviour based cohorts, that don’t reveal personal information. These would be updated regularly based on recent browsing behaviour. These cohorts are targeted, not the individual, and user data remains local, with the browser only sharing these cohorts with advertisers via API.

Over on GitHub, the FLoC proposal states:

“We plan to explore ways in which a browser can group together people with similar browsing habits, so that ad tech companies can observe the habits of large groups instead of the activity of individuals. Ad targeting could then be partly based on what group the person falls into.”

&

“Browsers would need a way to form clusters that are both useful and private: Useful by collecting people with similar enough interests and producing labels suitable for machine learning, and private by forming large clusters that don’t reveal information that’s too personal, when the clusters are created, or when they are used.”

Moving Quickly

It sounds like Google will be moving ahead with this pretty quickly, announcing that the plan is to start testing FLoC with advertisers in Q2 of 2021. 

“Our tests of FLoC to reach in-market and affinity Google Audiences show that advertisers can expect to see at least 95% of the conversions per dollar spent when compared to cookie-based advertising”

Expect to hear much more about this coming to Chrome over the coming months.