We take a look at Google’s offering to marketers as the chosen alternative to third-party cookies. What is Google trying to achieve with FLoC? How does it help marketers? Will marketers be able to shrug off the age-old tradition of working with third-party data and adopt data hygiene in their process?
Cookie – Thou Shall Be Missed
The third-party cookie is nearing its expiry date and the requirement to adopt more privacy abiding systems seems to have put pressure on marketers. Last year, when Google announced that third-party cookies wouldn’t be supported on chrome browsers, marketers were sent into a tizzy. In a hyper-personalised world, marketers walk the tightrope between customer data collection and abiding by data privacy laws. Google’s decision to do away with third-party cookies arises from the fact that the data protection laws globally have been created, and some reintroduced with more teeth. Violations under regulations like the GDPR can result in fines in millions. Plus, damages to affected parties. Also, Google knew it had to project itself as a responsible internet giant that cared for data privacy and also help marketers.
Introducing Google’s Privacy Sandbox
As an alternative to the third-party cookie, Google’s project privacy sandbox involves replacing the cookies with application programming interfaces (API) that receive only aggregated data about specific performance metrics like conversion. With a push towards collecting aggregated data, Google will rely on anonymised signals collected from chrome users. The API collects data for the ad industry from the users’ browsing habits without giving away their individual identity.
Ad targeting, measurement, and increased security are some of the goals that the privacy sandbox sets out to achieve. One of these APIs uses a machine learning framework that replaces cookies.
FLoC Flies Out Of The Sandbox
Google has been consistent in its communication and made its stance on data privacy pretty clear. In January 2021, Google spoke about building a privacy-first future for web advertising. In the blog, Google speaks about creating groups of people with similar interests, or characteristics, and collecting their data ensuring no instance of individual identity is given out to the marketer.
Google says this interest-based data collection will help marketers deliver advertisers to their customers based on broad categories of interests. Each category is called a “cohort”. With Federated Learning of Cohorts (FLoC), Google is essentially creating customer clusters from where marketers can collect data and deliver ads. Google thinks that cohorts can help erase personal identities. No personal data, no problem.
Interestingly, there’s no clarity on whether the user wants to share their browsing and other user data, albeit anonymously.
How FLoC Works?
Google’s FLoC is essentially a browser extension that looks like a CAPTCHA that has to be filled once by a web user. Google’s API will then assume that the user will continue to be a human in the future through trust certificates. FLoC works by using data as a signal for a browser’s activity. Google applies machine learning (ML) to create a cluster estimation technique that uses characteristics or shared traits from pre-determined customer sets. The estimates classify the data as shared traits based on the FLoC’s interpretation of given data parameters. Cluster analysis has been used to peak potential by Google while creating FLoCs. This technique allows Google to receive behavioural data from cohorts. The final analysis of a cohort’s characteristics can be used to match to marketing media without requiring to identify a user’s identifiable characteristics. The resultant data applies to every user classified in the cohort. This is akin to talking to a person hiding in a crowd. Google believes that it can uphold anonymity through the strength of the sheer number of users Google gets as traffic every day. Each marketer is given access to limited information from each cohort.
Also Read: First to Zero, The Perfect Data Journey
How Is It Different From Cookies?
Cookies traditionally help browsers remember users’ method of accessing the website and serves as a gate pass for the user should they want to revisit the website again. Cookies are text files that get an individual ID, unique to the browser, making it easier to identify the individual user.
While rolling out FLoC as a trial, Google said that it is a new approach to interest-based advertising that improves privacy and gives publishers a tool they need for viable advertising business models.
Google aims to help marketers keep growing their businesses and keep the web sustainable, with universal access to content.
Google Being Marketer’s Friend or Something Bigger is Cooking?
The cohorts have opened up the possibility of creating a universal identifier for end users. Most identifiers rely on tying third-party cookies to a centralised information database to create an ID that’s based not on details about a person but on the probability that a user belongs to a certain category. Most marketers use Google Adwords in their marketing campaigns. They are worried about the possibility of Google testing FLoC in the market now, but what if it decides to use the technology only for its clients or marketing teams. Google is known to support its users and it’s confirmed that marketers and ad vendors working closely with Google will benefit most when the third-party cookie becomes defunct.
Google’s third-party alternative to cookies is still undergoing multiple R&D sessions. Google has espoused the virtues of privacy and user identity protection through its statements multiple times. The internet giant shared statistics from a Pew Report that said 72 per cent of people feel that almost all of what they do online is being tracked by advertisers, technology firms or other companies, and 81 per cent think that the potential risks they face because of data collection outweigh the benefits. How well it manages to stay true to its promise of keeping the internet free and fair for marketers and customers, time will tell.