Last year, Google confirmed that a long-rumored change to the digital advertising community would be coming. In 2022, the Chrome browser will phase out support for third-party (3p) cookies. These cookies, without exaggeration, currently power the commercial Internet.
If you’ve looked at a pair of running shoes on Amazon, and then you see an ad for the exact same pair of shoes in the same size on another site later in the day, a third-party cookie made that possible.
At MLive and our other Advance properties, we use 3p cookies to better understand what content readers value, to improve their user experience, and yes, to collect data that assists our advertising partners to ensure they are targeting ads to the right person at the right time, whether they are on our site or other sites beyond our owned network.
That said, as Internet users, many of us are concerned about privacy, and over time, 3p cookie-tracking has become increasingly invasive. It’s time for the industry to move to a more privacy-centric approach – and you should be proactive in your planning for how you will face these paramount changes in tech.
On July 29, 2021, Jeff Sutton, VP of Programmatic for Advance Local, presented this topic. Providing relevant updates since the writing of this blog, many of the concepts in this article are described in this 20 minute session.
"It’s time for the industry to move to a more privacy-centric approach – and you should be proactive in your planning for how you will face these paramount changes in tech.”
Even if it is for the best for Internet users, the removal of 3p cookies is a highly disruptive change. We’re going to take a few minutes to share how we believe this will impact digital advertising, and how advertisers and media companies such as our own should prepare for this future. While anyone in digital marketing obviously sees this as a surmountable challenge, we also view it as an opportunity to continue to evolve and grow.
A Brief History of Third-Party Cookies
As the commercial potential of the Internet became apparent back in the early ‘90s, the biggest obstacle was that browsers were dumb – they couldn’t remember readers from visit to visit. Cookies were invented by legendary Netscape programmer Lou Montulli in 1994, who realized he could set a small piece of code in the browser that would recognize a reader when they returned to the site. This simple idea made browsers more than page readers, and when it was quickly followed by a lightning-fast search engine developed by a couple of students at Stanford University (Google), the modern internet was born.
Today, there are two primary types of cookies. First-party (1p) cookies are set by publishers and are used exclusively by those same publishers. Taking MLive or Cleveland.com as an example, they are used to recognize return visitors and to learn what they care about when they’re on the site. For example, if a reader visits a sports section three times in a week, they are considered to be a sports enthusiast – it’s pretty simple. They also power the tools that sites use to manage logins, identify readers as candidates to become registered users and then subscribers and more.
Third-party cookies are generated by publisher partners when readers visit their sites. These cookies allow companies such as Google, Facebook, Lotame (our own data management platform) and others to follow readers across sites. These “cross-site trackers” make the kind of data-driven audience targeting possible, and it’s how that ad for new shoes “follows” you across sites. They also enable campaign frequency capping and conversion attribution to the benefit of both buyers and sellers. The downside, though, is that they collect massive amounts of user data. Privacy legislation and regulations are generally moving away from this kind of unchecked data collection as it can potentially be tied back to an individual.
What Will Replace Third-Party Cookies?
Google wants ad targeting, measurement, and fraud prevention to happen according to the standards outlined in its Privacy Sandbox proposal. In this view, 3p cookies would be replaced by a series of APIs (application programming interfaces) that moves user behavioral data into the user’s browser, where it will be aggregated, anonymized, stored and processed to be used as targeting data.
The audience targeting side of this is called FLoC – the “Federated Learning of Cohorts.” It proposes to use machine learning to group people based on their common browsing behavior and inferred/observed interests, as an alternative to 3p cookies. In simple terms, this would move targeting and attribution of user behavior away from the publisher community and toward Google, and they report that early learnings are positive. As you can imagine, that has not been met with universal praise from the content publishers who have more control over targeting data today.
Additionally, cookie changes aren’t just limited to Google/Chrome. Apple’s Safari browser has been limiting its tracking opportunities, too, and since sites with massive reach, such as MLive.com, generate significant activity through that browser, we have continued to develop non-cookie-based strategies for that browser as well.
2021 will be the year when this all settles out, and it’s why we are so focused on preparing for broad changes to how we operate in this environment.
How MLive is preparing for these changes:
Across all of our Advance Local publications nationwide, we are implementing a Customer Data Platform (CDP) to enable us to think differently about our own site users. From an advertising perspective, we see our readers today as browser uniques, page views and impressions. The CDP lays the foundation for building a deeper first-party relationship with our readers, where we see them more as individuals and members of audience cohorts. We’ll enrich our understanding of them with unique 1p data that goes well beyond what we can do with 1p data today, and we will use that data to build audience targeting segments that give our technology as significant advantage in targeting accuracy and campaign performance.
In addition, we have gone all-in on identity match solutions that increase the value of the inventory we serve. The cookie has always been an imperfect proxy for the identity of the reader; our internal advertising, audience, and operations teams are working closely together to map out a customer journey that gives users an incentive to willingly share more identifiable information about themselves – an email address, a user account, a subscriber. In turn, the data is completely anonymized and passed along to a data match partner who enables us to generate increases in the value of every impression we serve, without 3p cookies.
Media companies are diversifying away from cookie-based advertising, where possible. Since changes in cookie tracking affects the entire industry, alternative ad targeting strategies are coming to market. In addition to cookie-based advertising, we’re working on cohort-based advertising strategies, contextual targeting strategies and, ironically, we’re seeing a resurgence in simple geo-targeting strategies – sometimes less really is more.
What Happens Next?
A challenge like this requires collaboration across various internal departments from advertising, subscriptions, analytics, and data, to test and prove the strategies needed to grow our first-party audience, and to develop the technical infrastructure needed to deliver next-gen audience targeting solutions. At MLive and other advance properties, this work has been underway for some time, as the writing was on the wall in regard to evolving privacy measures.
As a leader in the digital advertising space, not just locally but also nationally, we will keep the information coming – this is the first of four posts in a series; stay tuned for future updates on Identity, Targeting, Data and more.
Thanks for reading, and please reach out if have any questions about this rather complex subject that we may be able to help answer.
You can’t spell personalization without persona
In marketing, we hear a lot about personas and using them for targeting. Personas provide a way to view a group based on a collection…
Google Analytics 4: What You Need to Know
Debuting in the fall of 2012, the current version of Google Analytics, known as Universal Analytics (UA) has been the standard bearer of website data…
Match Expectations of Gen Z In Higher Ed
There is an urgent need to reinvest to make the academic experience of the younger generation flawless. But it’s not just about convenience. It’s also…