Google Analytics 4: What Cookieless Tracking Means

Google Analytics Logo
By
John Huggins

There are numerous changes that come along with the Google Analytics 4 (GA4) update. From relatively small changes, like the loss of bounce rate reporting, to more significant changes like data streams reporting being front and center. However you look at it, GA4 is different from what’s come before and it will shape the future of data privacy and digital marketing.

What is cookieless tracking?

GA4 has been created with a focus on privacy, and it can be used with or without cookies. With an emphasis on using machine learning and statistical modeling, GA4 is supposed to fill in gaps in data as the world shifts away from traditional tracking cookies for both privacy and security reasons.

What are Cookies exactly?

A cookie is a small file that stores a tiny piece of user data. Cookie origins can be traced back to the mid 1990s when their first documented uses were in e-commerce shopping carts.

These ‘bite-sized bytes of code’ (pun intended), or cookies, can save a variety of different information, depending on what the website owners want to track.

Tracking cookies also allow remarketing campaigns to follow your activity across the internet, which is one of the main concerns for privacy advocates. As you visit more and more websites, and accumulate more data, cookies can put together a more detailed picture of who you are, what you like and dislike, and even predict behavior–especially purchasing behavior.

That kind of predictive purchasing behavior analysis has a lot of value in the marketing world but also presents significant potential for abuse.

Are Cookies Good or Bad?

Everything in moderation. Cookies aren’t inherently bad, and in several ways, are very useful. Cookies help create a personalized experience for site visitors. The auto-fill option when checking out has been touted as extremely helpful. For trustworthy sites, there’s nothing wrong with allowing cookies.

But like their yummy real-life baked counterparts, it's very difficult to figure out where to draw the line. Privacy-conscious individuals are concerned about the potential for cookies to allow access to sensitive personal information. In response to this, the European Union created the General Data Protection Regulation (GDPR) law in 2018 which allows website users to choose whether or not they want to accept cookies on their device. And the cookie consent pop-up banner was born.

Third Party Cookies vs First Party Cookies

Cookies are broken into two categories: third-party and first-party.

The primary issue for most users isn’t whether a cookie remembers what’s in your shopping cart, a lot of busy people live and die by what their trusty cart remembers for them. The real issue is when businesses track you across multiple websites and create a tapestry of your personal data for remarketing. That cross platform tracking is made possible by third-party cookies. The data collected about your activity, interests and general online habits, like frequently visited news sites, are then used to create demographic profiles for ad tech companies. And for many users, this level of insight into personal activity on websites and apps, is the main point of contention.

First-party cookies, unlike the hated third-party cookie, are only used by the website that the user is actively visiting.Google Analytics 4 uses first-party cookies in order to track data and remain compliant with regulations like GDPR and the California Consumer Privacy Act (CCPA). These cookies are seen as more acceptable and are what allow users to keep their passwords stored or their cart contents active.

These privacy developments are also being seen in tech manufacturers, like Apple's iOS14, which confirms the future is likely cookieless and folks need to get on board. A more private internet environment sounds great, but how can just first-party cookies really compare to the all-intrusive third-party cookie?
Google’s answer is machine learning, or Federated Learning of Cohorts (FLoC), that uses blended data.

FLoC - A Simulated Interest-Based Audience

With the current rise in AI, the Federated Learning of Cohorts fits right in - it's definitely a work in progress but it's a huge part of the cookieless browsing future. With FLoC, Google simulates user data rather than using third-party cookies and Google is incredibly confident in its ability, claiming in their blog:

“Results indicate that when it comes to generating interest-based audiences, FLoC can provide an effective replacement signal for third-party cookies. Our test of FLoC to reach in-market affinity and affinity Google Audiences show that advertisers can expect to see at least 95% of conversions per dollar spent when compared to cookie-based advertising. The specific result depends on the strength of the clustering algorithm that FLoC uses and the type of audience being reached.”

So in translation, FLoC was created to provide a balance between protecting privacy and allowing for ad selection based on interests. When a user is browsing the web, the browser (i.e. Chrome) will use the FLoC algorithm to group them into a cohort with other people who have comparable first-party cookie histories.

A FLoC of Problems

FLoC is an intriguing attempt to address online privacy concerns, but it is not as effective as desired in terms of digital marketing. For instance, Google, rather than the advertiser, decides the interest cohorts.

 As a digital marketer/advertiser, you have to fit your product or service into Google's pre-made categories and immediately lose a significant amount of metrics used for targeting. And, frustratingly, it doesn't end there.

Clarity and attribution will also be lost. Online marketers will no longer be able to track users across different websites OR between devices. The marketing attribution models will become more difficult when mapping a customer journey that started on a personal mobile device and ended on a workplace desktop.

And ultimately, FLoC puts most of the power into Google’s black box. It's unlikely we will ever have a complete understanding of all the parameters that make up an interest cohort, which just further reduces transparency and almost completely negates any offline audience data analysis. Think you know your audience? Google may disagree

What’s Next

The world of digital marketing is always evolving and GA4 seems to be ushering in a new era that will fundamentally change marketing online. But it’s not all doom and gloom. If the Internet has shown us anything, it’s that change is constant. Eventually, new tactics and strategies, and likely new technologies, will find better ways to market online and also maintain privacy. And that’s a win for everyone.