Are Businesses Ready to Do Without Third-Party Tracking and Can Generative AI Help?

Later this year Google will fully deprecate third-party cookie tracking of all users. Why is this such a big deal? Google is responsible for about 82% of the world’s search traffic, its Chrome browser boasts a market share in excess of 60%, and its digital media ad server and YouTube ad products are far and away the leading digital advertising products in their respective markets.

Google’s phase-out of third-party cookies represents a nail in the coffin of advertising measurement tracking with this standard approach. In addition to the world’s largest advertisers, publishers, and broadcasters, the implications of this move have been watched for many years by investors, national governments, trade and competition commissions, and justice departments. The effects on all promise to be resounding.

Every marketer expects this to rock their world. The question is, are they ready? And if not, how can they compensate for the loss of easy access to the de facto source of customer behavior prediction to inform their performance marketing decisions? How can marketing effectiveness be measured without the data that third-party cookies so conveniently provide? Is there a role for generative AI?

In preparation for this monumental shift, last year Google released an attribution measurement reporting solution as part of its privacy sandbox product for advertisers as their alternative to third-party cookie measurement. The solution, a data application programming interface (API), gives the ad ecosystem access to a combination of ad log level of individual ads served and an aggregate record of ads served. Together these data sources provide a comprehensive and representative measurement of advertising activity purchased by advertisers and served to the market, as well as more detailed information about the sequence of ads served to individual users at the event level.

So how does a marketer blend that data to recover the full picture of the customer journey to determine return on advertising investment in an effort to optimize campaign execution? While the term generative AI typically refers to language and imagery generation we shouldn’t forget that numbers underlie the predictive renderings of the models that generate language and imagery. So ask yourself, why not use powerful predictions to render the consumer data being lost?

In a recent study survey conducted by Statista*, a prominent global marketing research firm, 200 CMOs from six industries were asked for their thoughts on the impact and importance of AI as a strategic marketing planning and measurement tool.

The survey results indicated that 75% of CMOs agreed that AI’s can generate profound detail into customer behavior; 73% agreed that real-time data processing and interpretation by AI facilitates agile and informed decision-making; and 70% agreed that AI can be used to predict the effectiveness of creative imagery and messaging. Yet the headline is that 75% have not yet taken steps to prepare for measurement once the digital cookie jar is empty.

While this may be surprising at first glance, a disciplined approach to adopting new technologies is the best practice. Depending on the nature of the marketer’s business and experience with the adoption of transformative innovations, the journey and where you as a marketer are expected to be on it widely differs. But one thing is sure, putting off the journey is the worst decision of all.

Third-party cookies have only ever captured a distorted and disconnected picture of the value of a media investment. Marketers were never even close to getting a full and representative picture from third-party cookie data even in the heyday of their efficacy. Privacy-preserving generative AI synthesizes the data you need to manage and report on your business, with the data you have. It provides a deeper understanding of the customer and the investment in that customer, leading to a different kind of decision-making directly connected to marketing investment impact. It bridges the divide between the Chief Marketing Officer’s operation and the way the rest of a business manages risk and reports on investment returns. This synthetic data set is more real, manager-acceptable, and comprehensive than deterministic and personally identifiable data ever was. This type of data synthesis that can be done using generative AI on the type of data Google now provides, consolidated with other first and third party data, is commonly referred to as data augmentation.

As a strategy to address challenges related to data accessibility, complexity, and availability, the use of synthetic data has seen remarkable growth, with Gartner forecasting its adoption could reach 60% by the close of 2024, a stark increase from just 1% in 2021. Gartner also predicts that by 2030, synthetic data will overtake actual data in training complementary generative AI models, such as those AIs used to have conversations with the data.

For measuring communication effectiveness to potential customers across the sales funnel, across sales channels, and across paid, owned, and earned media channels in a privacy- preserving way allows marketing managers to harmoniously manage brand and performance marketing alongside other strategic business imperatives such as sustainability and corporate social responsibility.

We may be losing cookies, but with a plan to commence the transformative journey that generative AI will take us on, and with the innovative methods that marketers have now been using for a couple of years, we are gaining the opportunity to rethink measurability and empower managers with massively better judgment. That could not have happened with reliance on third-party cookies.

-Michael Cohen. Chief Data & Analytics Officer, Plus Company

_{*[Disclosure: This research was commissioned by Plus Company, but it was conducted independently by Statista. The data is available upon request.]}