The age of third-party cookies is coming to an end. Browser cookies — the technology that has enabled marketers to deliver targeted, relevant internet advertisements to consumers – are on their way out, with Google closing the deal by phasing them out entirely in Chrome by 2023.
Instead, in response to a slew of data-privacy concerns plaguing digital advertising for years, Google is testing the Federated Learning of Cohorts (FLoC).
If you’ve ever run advertising on your website, you’ve most likely used third-party cookies. Even if you are the one who places advertisements on publisher sites, removing cookies and replacing them with Google FLoC in Chrome will result in long-term changes to how you handle online advertising.
After all, Google Chrome is likely to be used by a large portion of your target audience. Right now, it owns 46.81 percent of the overall browser market share in the United States, with the nearest competitor, Apple Safari, following closely with 37.72 percent.
WHAT IS GOOGLE FLOC?
Federated Learning of Cohorts (FLoC) is a new approach for businesses to reach potential clients with relevant ads by grouping together big groups of people with similar interests.
This provides an additional degree of privacy because data is kept locally on users’ machines rather than on servers, while also allowing marketers and publishers to target cohorts.
Here is Google’s definition of cohorts: “Groups of people with similar interests (might) replace individual identifiers” (2021).
In other words, the Google FLoC is the tech giant’s effort at a privacy-first solution that shields user data from monitoring while still allowing marketers to access the web. It’s an API (application programming interface) derived from the Privacy Sandbox protocol.
In short, FLoC provides a privacy-preserving mechanism for interest-based ad selection
WHAT WOULD FLOC BE USED FOR?
- Display adverts to persons whose browsers have been identified to often visit an advertiser’s site or to have an interest in relevant themes.
- In order to influence ad auction bidding behavior, machine learning models are used to estimate the likelihood of a user converting based on their cohort.
- Content should be recommended to users. Assume a news site notices that their sports podcast page is particularly popular with visitors from cohorts 1234 and 7. They would then have the ability to recommend such material to other visitors from those cohorts.
HOW WILL GOOGLE FLOC AFFECT DIGITAL ADVERTISING?
To begin, you should be aware that Google FloC is a technique that allows browsers to enable center-targeted interest.
This method works by collecting data on users’ browsing behaviors and then grouping similar people into cohorts based on their habits and interests.
Depending on the individual Google FLoC Proposal on Github, the algorithm responsible for creating these cohorts also evaluates the visited site URLs and the content of the viewed pages.
The cohort information may then be obtained by advertisers for a variety of digital marketing objectives. Individual user data, on the other hand, stay local, in the browser, with the latter revealing just the cohort ID.
Furthermore, a cohort would be made up of a bigger group of people, making it impossible to identify any of the individual users within the group.
At the same time, the data supplied is detailed enough to allow advertisers to target their ads effectively. Furthermore, each week, users are assigned to new cohorts depending on the various browsing statistics obtained the previous week.
Google had stated that it will begin testing the new technique with advertisers in the second quarter of this year. Google stated at the time that thus far, tests have yielded encouraging results, and marketers should expect to see about 95% of conversions for dollars spent vs cookie-based advertising.
Experts believe that this is a significant shift in the digital game. For starters, advertisers are accustomed to utilizing cookies to target specific persons with their advertisements. Google FloC alters this by eliminating the individual and replacing it with cohorts formed based on shared interests. This offers an additional degree of anonymity, which can aid in addressing the growing issue of user privacy.
Another significant distinction is that cohort assignment occurs in the browser, which implies that the user information remains local, in the browser.
WHAT DOES GOOGLE FLOC DO FOR ADVERTISERS AND CONSUMERS?
The primary impact of Google FLoC replacing third-party cookies will be on marketers. To continue monitoring, they may have to accept Google FLoC or accept that advertising cannot be as precisely targeted as it once was.
The extent to which internet consumers are concerned about FLoC will be determined mainly by their view of targeted advertising and data sharing. Some online firms would undoubtedly suffer if they made money from tailored advertising, however, publishers will still be permitted to employ first-party cookies.
We’re getting to the stage where we need to talk about why we claim the internet is “free.” It was never free! Many internet users will undoubtedly continue to gladly give up their privacy in exchange for “free” access to content. The rest of the time, Google FLoC may be easily avoided.
The most straightforward approach to avoid Google FLoC is to use a browser other than Google Chrome.
Furthermore, Chrome users may disable the technique by disabling third-party cookies in their browser settings. Web developers can also opt out of Google FLoC. Because they employ first-party cookies rather than third-party cookies, mainstream online apps are unlikely to be immediately affected.
THE PROBLEMS WITH GOOGLE FLOC
You may have also heard that Safari and Firefox have discontinued support for third-party cookies due to increasing privacy concerns. Experts are concerned that although FLoC’s grouping approach may be insufficient to prevent hackers from gaining access to individual user data.
Previously, specialists were concerned that Google FLoC would not be up to the task. Most marketers now agree that Google FLoC falls short of what most advertisers expect. And, despite the fact that the system is still under development, several experts have already identified its shortcomings.
For starters, many argue that it will never be as effective as third-party cookies since it only works on Chrome, leaving out crucial aspects such as cross-device, cross-browser, and offline data.
On the other side, others argue that the targeting is crude because the cohorts are predetermined and fully generated by Google. Measuring product interest in this manner is challenging, making remarketing nearly impossible. Furthermore, because cohorts have replaced identification, assessing campaign success becomes increasingly complex.
Other issues include the fact that Google FLoC will strengthen Google’s market dominance and that the technique protects privacy from everyone except Google. Google would still be able to access cohort data and raw user data saved in the browser cache using Google FLoC.
In addition, others say that FLoC will totally give Google the advantage in the digital world if it becomes the norm to replace third-party targeting techniques. Google will be free to modify the algorithms as it sees fit.
Finally, others believe that the approach will not improve since it is just not in Google’s best interests to make non-Google ad spending too competitive. Having too much data on the browser would very certainly pose significant security issues owing to the possibility of de-anonymization.
As you can see, there are a few solid aspects that instantly push the method’s advantages to the sidelines.
GOOGLE FLOC AND ITS PRIVACY RED FLAGS
By eliminating the need to store online users’ information on servers, privacy should improve. Additionally, Google contends that FLoC provides greater privacy protections than third-party cookies or alternatives such as browser fingerprinting. Google refers to FloC as a “privacy-preserving API,” in part because marketers only have access to the cohort ID, not individual user identities.
However, when cohorts contract – or, in advertising parlance, become more focused – the potential of unintended identification rises. “If I am a trader who specializes in performance motorcycles, anybody who visits my site will be assigned into a cohort based in part on their interest in performance motorcycles.”
If cohorts are formed based on limited geographical regions or other connections, such as to an employer, the privacy risk increases even more. Some interests will be excluded from cohorts, such as pornographic websites and medical information.
However FLoC classifies these interests as ‘sensitive,’ and the system may be unable to discern between a history of accessing adult content and studying, say, Covid-19 symptoms. A website owner that has access to their customers’ personally identifiable information might potentially utilize such data to associate cohorts with people.
It is unclear how FLoC’s cohorts would operate in reality, but Google employees have stated that the system will not be tested in the EU because of fears that it violates portions of the GDPR and the ePrivacy Directive. FLoC is now being tested in Australia, Brazil, Canada, India, Indonesia, Japan, Mexico, New Zealand, the Philippines, and the United States.