Targeting As We Know It

In the dynamic landscape of modern advertising, the advent of Big Data triggered a seismic shift for businesses and consumers alike. Nearly two decades later, however, marketers find themselves in a paradoxical situation, grappling not with scarcity but with an abundance of data. The challenge has transformed from acquiring information to navigating the complexities of honing in on the most pertinent data. As cookies face extinction and walled gardens become increasingly impervious, marketers are confronted with the task of forging meaningful connections with their qualified audiences while also measuring the effectiveness of these connections.

As the traditional tools for data gathering undergo a transformative phase, brands are compelled to reassess their strategies for engaging with consumers. The diminishing relevance of cookies and the fortified barriers surrounding walled gardens (closed ecosystem or platforms that limits external access and data sharing while offering a controlled environment for audience segmentation) pose a pivotal question for marketers: How can they authentically engage with their most valuable audiences in an environment where conventional digital marketing avenues are disappearing? Beyond mere engagement, there’s the pressing concern of gauging the impact of these interactions on brand success. Following today’s cookieless future, the narrative shifts from the abundant pursuit of data to the nuanced art of extracting value from the bandwidth of available data, challenging marketers to innovate and redefine their approach to audience targeting and performance measurement. In this blog we’ll discuss our POV on targeting and measuring in the new cookieless era and the impact it will make on your current strategy.

Top Targeting Solutions That Don’t Rely on Cookies

The imminent demise of Third-Party cookies poses a substantial threat to the digital advertising landscape, with alarming implications for publishers. Research conducted by the Interactive Advertising Bureau (IAB) underscores the gravity of the situation, suggesting that publishers could face a monumental loss of approximately $10 billion in ad revenues if these cookies are eliminated. The magnitude of this financial setback is further emphasized by Google’s stark warning that publishers risk forfeiting a substantial portion—between 50% to 70%—of their revenue if they fail to adopt a new approach toward user data. The stakes are high, as echoed by a resounding 77% of marketers who express a shared concern that Google’s decision to phase out Third-Party Cookies will undoubtedly amplify the challenges of their marketing endeavors, creating a landscape where navigating and optimizing advertising strategies becomes notably more intricate.

That being said, there are multiple alternatives to third-party cookies that marketers can leverage to say connected to consumers in the digital landscape:

• First-Party Data: First-party data, sourced from various customer interactions including websites, surveys, email, and CRM systems, often comprises basic information, but building trusting relationships and obtaining user consent can enable the collection of more detailed data through progressive profiling strategies.
• Google Topics API: Coming in 2024, Google Topics API or Privacy Sandbox categorizes visited sites into topics for displaying relevant ads, ensuring user privacy by not sharing specific site information and allowing users to choose their associated topics.
• Universal Identifiers: Unique universal identifiers enables brands to identify users across websites and devices, providing benefits such as cross-device tracking, a seamless experience, data loss reduction, and improved accuracy in sample sizes, and can be crafted using First-Party Data, balancing targeting and privacy.
• Fingerprinting: Digital fingerprinting, comprised of Device Fingerprinting and Browser Fingerprinting, collects device-related information like plugins, IP, browser, operating system, screen size, and time zone to create unique identifiers for users based on their internet browsing and installed third-party apps
• Contextual Targeting: Serving ads on web pages by analyzing the content keywords and phrases, without relying on personal data. Publishers may sometimes share device and browsing time details. Our recommendation is to use a DSP, such as DV360, that uses machine learning to predict suitable pages and timing for their ads, with the most success occurring on pages with highly themed content that aligns with specific user interests.

Contextual is the New Behavioral Targeting

Contextual targeting in programmatic advertising tailors ads to users based on the content they engage with, analyzing attributes like keywords or themes on a webpage. By aligning ads directly with the viewed content, contextual targeting enhances the potential for engagement, providing brands with a tool to establish more meaningful and personalized connections with their audience.

Contextual advertising is most effective when displayed on relevant and recent content, especially given that news articles constitute a significant portion of new online material. With 74% of news articles reaching peak traffic on their publication day and an additional 25% peaking the day after, the short lifespan of such content highlights the importance of timely ad placement (note: ensure that your contextual strategy is aligned with machine learning/AI features in your digital platforms).

Contextual targeting addresses this by assessing various content factors including:  

• URL domain
• Keywords
• Category
• Semantics
• Meta tags and
• Bidstream signals

These attributes, hand in hand with AI, allow marketers to determine both the relevance and recency of content, to align ads more effectively with the context of the user’s interests. For instance, keyword targeting matches ads with specific keywords in the viewed content, while category targeting places ads based on broader content categories like beauty, automotive, or finance.

Marketers, Meet Your New Bestie – Probabilistic Models

Now that we’ve covered targeting solutions, how will measurement and additional scaling be affected by this new cookieless era? Currently there are two types of data measurements: deterministic data and probabilistic data. In today’s digital landscape most marketers use deterministic data since it leverages both pre and post consumer interaction with your ads on publisher sites or apps. However, deterministic data does not align with robust privacy practices or the cookieless era forthcoming.

By using contextual data and predictive audiences, marketers can lean into probabilistic models to fill in the gap (data that is currently available in programmatic marketplaces). These datasets, which don’t rely on cookies or device-based identifiers, offer the ability to target consumers based on inferred interests and anticipated responses, all while maintaining user privacy. This shift allows advertisers to construct cohorts or employ vector-based audience targeting on a broad scale, offering an alternative that sidesteps the privacy concerns associated with deterministic data.

Futhermore, the future of advertising relies on adopting innovative models, especially privacy-safe probabilistic data, strategically applied on the supply side. Currently, the dominant targeting method relies on third-party cookie data at the DSP level for open exchange inventory. However, impending cookie deprecation poses a significant threat, potentially leading to an 80%-90% loss in targeting efficacy. The envisioned future involves applying cookieless data directly on inventory at the SSP level, addressing data privacy concerns and streamlining the supply path. This next generation of probabilistic data offers advertisers sought-after addressability, scale, and performance in a privacy-safe and optimized landscape, particularly in private marketplaces driven by deal IDs.