Why Cookieless Continues to be the Right Strategy

Keep in mind that Safari and Firefox have not supported cookies since 2019 and some Chrome users had already chosen to block cookies, leading to 51% of internet traffic already being cookieless by default.

Additionally, more money continues to flow to natively cookieless environments like CTV and OOH. The number of consumers who are primarily reachable through cookieless targeting will continue to grow. Clearly, no advertiser should write off more than 50% of consumers, so they need to make cookieless a major component of existing and future strategy.

The Cookieless Infrastructure is Here

While Google’s about-face may not be completely surprising, many advertisers and publishers proceeded as if it was definitely going to happen. At last count over 50 alternative identity solutions are available in Prebid alone.

Going cookieless doesn’t mean advertisers can’t target anymore; they just have to do it differently. Consumers and advertisers alike have no interest in returning to the days of untargeted ads, luckily there are plenty of options to choose from.

Adtech partners have the opportunity to step in to enable strong targeting and measurement in a cookieless environment.

First-Party Data Only Goes so Far

While third-party cookies are going away, first-party cookies remain an important part of modern advertising. Publishers request data from visitors when they subscribe to newsletters or access content. They can use them to target consumers better and provide measurement to advertisers.

First-party data is why one of the hottest trends in advertising is the retail media network. Retailers are sitting on an incredible amount of customer data, such as purchase history, location, preferences, and other requested identifying customer information. The first iteration of retail media networks involved retailers offering real estate to advertisers on their website to target potential customers, but it has quickly expanded to extended networks where retail media data can power ads on third-party sites. While this is still only a percentage of the overall digital ad spend, we need solutions for targeting elsewhere.

The Key to the Future is Alternative Identifiers

The Trade Desk has created a global identity standard for the advertising industry with Unified ID 2.0 (UID2), an open-source authenticated framework that uses encrypted email and phone number data. But this alone won’t solve the cookieless targeting issue. As a standalone solution, it lacks the scale to reach the larger population of non-logged-in traffic.

In addition to authenticated solutions, there are a couple of ways to bring successful targeting to cookieless through alternative identifiers:

• First-party Cookies and Data: Data collected directly from a marketer’s audience helps them identify and target their audiences. Retailers particularly benefit since they sit on a significant amount of customer information.
• Probabilistic: Authenticated solutions will act as “seeds” for modeled expansion for non-authenticated users. Probabilistic targeting uses soft signals to build a user profile such as browser type, device type, and other non-identifiable characteristics.
• Contextual Advertising: Contextual targeting infers and delivers ads based on consumed content. For example, a visitor reading an article about a fashion designer on a news site is likely either a fan of that fashion house or fashion in general. The information surrounding the ad creates signals that programmatic systems can use to place the right ads.

The disappearance of the third-party cookie may not come as quickly as anticipated, but it’s safe to assume more and more consumers will opt out of tracking. There is an obvious first-mover advantage in embracing the future and focusing on today’s cookieless environment.

It’s essential to evaluate key factors that ensure both effectiveness and privacy. Prioritizing transparency in how identifiers are used between buyers and sellers is crucial, whether through deterministic, probabilistic, or contextual methods.