Google FloC Explained: A Guide for Marketers

As FLoC testing is officially underway, we’ve compiled everything that we know about FLoC: what it is, how it’s different, and what advertisers can do to prepare.

What is FLoC and Why is This Change Happening?

Privacy continues to be a big issue as it pertains to digital advertising. Ad platforms and browsers alike are taking steps to improve transparency and privacy. Other browsers had taken privacy protection steps and this is now Google’s way of responding to those same concerns.

Google had announced that they would no longer support third-party cookies but also saw a need to create advertising solutions with the goal of circumventing ad platforms from developing other, more invasive ways of tracking – such as fingerprinting. The outcome of those efforts were Federated Learning of Cohorts, or “FLoC”, for short.

With FLoC, instead of individual cookies being aggregated at the individual level and passed from sites to platforms, browsing data would live within each individual browser. FLoC would then assign individual browsers to larger groups (or cohorts) of thousands of users based upon that browsing data.

“Cohorts are dynamic and will update every seven days during the initial trial.

As a person’s browsing behavior changes, their browser will assign them to a different FLoC cohort that reflects those interests.

For example, at one point they might be in a FLoC cohort with thousands of other people who have also recently visited websites about gardening and travel overseas, and then at another point in time they could be in a group of people who have recently visited sites about art supplies and cooking.” – Ginny Marvin, Google Ad Products Liaison

How Is This Different Than the Previous Process?

The key takeaway from a privacy standpoint is that users’ individual browsing history therefore is no longer shared.

Instead of publishers being able to ‘follow’ people as they jump from website to website, a person’s browser history is held by FLoC and isn’t shared with anybody, not even Google.

That data previously fueled targeting options for advertisers in the form of affinity audiences and in-market audiences and also contributed to conversion tracking so advertisers naturally have questions about what to expect and how to prepare.  Before we get into that, let’s cover some of the pushback that FLoC is receiving from the industry at large.

Why Is FLoC Getting a Bad Rap?

You may have read articles about other organizations pushing back on the idea of FLoC. Remember how I mentioned that Google was proposing FLoC as a mean to balance privacy concerns and advertiser needs?

Opposers suggest that FLoC creates new privacy concerns because it can still be paired with more invasive forms of tracking (such as fingerprinting) that it aimed to circumvent.

With that in mind, it’s possible that someone could continue down the path of more invasive tracking and then layer FLoC information on top of it and ultimately wind up with a much more granular view than before.

The other issue is that, although the browser will only share data with individual sites as part of a larger cohort, some of those sites will have the ability to identify specific users (for example, folks that log in or make purchases). Therefore, it will still be possible to learn individual user data and then use that data to make inferences about the larger cohort.  It’s possible that data could eventually be exchanged at a larger scale.

“The power to target is the power to discriminate. By definition, targeted ads allow advertisers to reach some kinds of people while excluding others. A targeting system may be used to decide who gets to see job postings or loan offers just as easily as it is to advertise shoes. “ – Bennett Cyphers, Electronic Frontier Foundation

While Google Ads does have policies and reviews in place within the ad manager platform to ensure that discrimination does not occur, because FLoC is a product of Chrome –  not Google Ads – it’s important to remember that it would be powering data for other ad managers so this is a worthy concern.

Cohort Eligibility – Who Can Be Tracked?

A cohort will not be eligible to be advertised to if it includes a history of visiting sites with sensitive topics at a high rate.

This seems to apply to websites that publishers were already restricted from using as part of personalized advertising.

Therefore, cohorts regularly visiting websites that fall into the following categories will not be eligible for publishers to access:

  • Personal hardships.
  • Identity and belief.
  • Sexual interest.
  • Access to opportunities

Public Release of FloC

The FloC-created cohorts will be available for public testing. Preliminary testing began in March, and Google announced that advertiser tests are now officially underway.

There is a whitepaper released for developers suggesting how to run your own simulations, which can be found here.

How Can Advertisers Prepare?

Google’s initial trials of FLoC found that it performed similarly to In-Market and Affinity audiences, so the hope is that advertisers will see similar performance to what they’re already used to.

“There is still work and testing to be done, but the aim is to continue to deliver advertiser outcomes and publisher monetization while prioritizing user privacy.” – Ginny Marvin, Google’s Ad Product Liaison

It’s important to remember, as well, that third-party cookies are going away but first-party cookies are not. As a quick refresher:

  • First-party cookies are those that are set by the current website that a user is visiting. This allows the domain that a visitor is currently browsing, to store analytics data and a host of other items that can benefit user experience, such as keeping a user logged in, or saving their cart even if they aren’t signed in.
  • Third-party cookies are set by (you guessed it) a third party, in order to track user behavior across sites. For instance, third-party cookies are used to track and deliver ads to folks as they navigate publisher sites on the Google Display Network.

First-party data is more valuable than ever. Advertisers would be wise to make efforts to collect as much first-party data as they can.

Google has recommended a few times now that advertisers should ensure that all websites are tagged with the Google Ads universal tag.

There is a max of 5 domains per first-party set, so it’s important that companies using several different vanity domains should consider consolidating.

First-Party Data Usage

Another question on the minds of advertisers: how will we be able to use our own first-party audiences?

Google has a proposal dubbed “FLEDGE” that incorporates industry feedback into how this would be handled. User data, including browser history, would remain shielded from buyers and sellers. Google notes it will be available for testing in origin trials later this year. Ad tech companies can test the API using a “bring your own server model.”

Though FLEDGE testing is still in progress, advertisers will have the ability to utilize customer lists in essentially all of the largest advertising platforms.

Google announced at its recent Google Marketing Livestream, that it would be lifting restrictions on which advertisers are eligible to use customer lists. Microsoft Ads also recently announced that advertisers will be eligible to target customer lists through the Microsoft Ads platform. Facebook, Linkedin, and other social platforms also offer similar options.

Conversion Tracking In a Third-Party-Cookieless World

While conversion tracking is largely based upon first-party data, third-party data does play a role as it pertains to cross-device and view-through tracking. Google is testing different ways to capture this data without utilizing third-party cookies.

The measurement proposals leverage an API that has the capability to report both event-level and aggregated information. With the goal of preserving privacy, the API uses the following techniques:

  • Aggregating data that is reported in order to ensure that each person’s activities and identity remain anonymous.
  • Limiting the amount of information reported about each conversion so that it isn’t possible to expose the identity of the person behind the conversion.
  • Adding “noise” to the data that is reported, to protect each person’s privacy by including some random data along with the actual data.

Google notes that it is beginning to run simulations in order to better understand how different use cases might be impacted by the various proposals and will be sharing findings in the future.

Get Involved in the Conversation

Google’s privacy sandbox website offers information about the initiatives it’s looking into to provide service users peace of mind regarding their privacy.

Current initiatives include:

  • Preventing tracking while browsing the internet.
  • Preserving the open web.
  • Helping publishers build sites that respect user privacy.

If you’re a publisher, advertiser, developer, or part of an ad tech company, you are invited to contribute to the discussion.

Despite the launch, FLoC is still in development and could evolve based on user experience and feedback.