With the ever-evolving ad industry toward privacy and security, publishers are seeking innovative ways to maximize their data monetization strategies. As it will soon be impossible to rely on traditional tracking technologies, the need for robust solutions to identify users becomes more sharp. Identity Solutions are stepping up to help publishers with user identification in a fragmented ecosystem.

In our previous article, we explored the concept of an ID Graph and its practical applications. An ID Graph is the result of the process known as Identity Resolution. To complete this process, publishers use a set of operations involving the collection, processing, and linking of IDs to establish unique user identity groups at different levels: individual, household, trait and event. In simple terms, it allows a publisher to determine or infer what individual is visiting its websites or apps, whatever device, email, or IP address he comes from, therefore targeting him with a relevant ad.

‍

CDPs Are Not Enough Anymore

Initially introduced in Customer Data Platforms (CDPs) and used by marketers, ID Graphs has seen adoption by publishers, reflecting the changing dynamics of the industry. The move toward first-party data and the necessity to introduce new advertising and monetization strategies have driven publishers to opt for new ways of building their audiences. Publishers have utilized CDPs to consolidate all available data, create user profiles to increase the value of their ad inventory for advertisers, and enhance targeting capabilities. However, the limitations of CDP capabilities in real-time data processing, basic identity resolution and segmentation are insufficient to support the complex digital ecosystem where users navigate today and also add additional cost to publishers monetization stack.

As a result, ID Graphs now transcend their original CDP scope in more intricate systems, evolving into complex solutions and integrating into platforms, where organizations can securely collaborate around user data and seamlessly activate it. This expansion unlocks new possibilities for publishers, including those in broadcasting, TV networks, and audio platforms, to monetize what they have on the table.

‍

Accurate Identification for Better Advertising

Optable’s ID Graph is an example of such an advanced solution, providing highly interoperable environment that unlocks many use cases within the sole platform such as audience segmentation, harnessing insights, audience activation, data collaboration, programmatic bidding with enriched IDs and Privacy Sandbox applications.

To construct an accurate and rich identity graph, Optable groups identities by running several critical operations during the resolving process: 

• Auto-normalization to identify the type of ID and cleanse it. For example, spaces are removed from email data, and the format is adjusted according to the specific ID requirements.
• Deduplication and resolution across all IDs, including resolving IPs which CDPs usually do not do.
• Backend hygiene to prevent over-linkages or extra large clusters, ensuring accuracy and delivering precise targeting
• Setting up the numbers of associations to control bid enrichment application, which is not a use case in CDPs
• Configuring the shape and size of the graph. A larger scale implies lower accuracy, similar to look-a-like models.

‍

‍

‍

Different Data Types for Improved Identification: Deterministic And Probabilistic Identifiers 

ID graphs are not made the same. They can be built and scaled differently depending on the type of data matching used. There are two following ways to do that:

• The deterministic method of matching is based on data explicitly given to users; therefore, those are identifiers that are most certainly linked to the individual: names, emails, and telephone numbers provided by the person. This approach is more accurate, but it is harder to scale as these data are more challenging to acquire or enrich. The alternative solution for the publisher here is to use the existing first-party data networks of deterministic identifiers. Examples include Experian’s LUID, TransUnion’s TUID and LiveRamp’s Ramp ID. While these centralized ID providers provide a strong foundation for creating an ID graph they oftentimes come with limitations on how they can be used and therefore should be thought of as part of our overall solution.

• The probabilistic method is a broader concept based on predicting user behavior and plausible events similar to those noted for similar IDs originating from identifiers on the household level. Probabilistic IDs are the product of analyzing and stitching different cross-channels and devices' data signals. The method is based on similarities and probabilities and, consequently, is not 100% accurate. However, one of its advantages includes the possibility of measuring a large number of probabilistic events and scaling the ID Graph faster.  

Companies like ID5 and Predactiv offer probabilistic IDs. These providers process signals such as device IDs, IP addresses, behavioral and contextual data to infer the person’s identity and increase data matching rates.  

These two methods created quite a buzz in the industry, arguing that deterministic matching is the right and only way to match data accurately. However, the answer lies in the golden balance, where two methods are combined in different proportions. Here, the publisher must find its own ratio between accuracy and scalability. 

‍

Enhancing Ad Targeting and Boosting Revenue

In the cookieless environment, with addressability being continuously undermined from signal, identifying and targeting users is an important goal. Publishers can significantly increase their revenue by using purpose-built identity resolution to create comprehensive identity graphs. There are several reasons for this.

First, consolidating and linking data points together on different levels allows publishers to identify and reach more individuals and households within desired customer groups. With a large number of identifiers and licensed data providers evolving in the market, publishers are also able to amplify ad addressability by enriching audiences and scaling graphs. 

Second, by injecting deterministic and/or probabilistic IDs into their database, media companies can activate programmatic ads and achieve higher bid density. Thanks to enriched audiences in the bid stream, publishers can increase revenue.

Third, by working with a third-party data provider, publishers can resolve their identity graph to partner datasets to create new addressable audience segments. These new segments, such as demographics like gender, age, or household income, can then be packaged and sold to advertising partners for ad activation.

Lastly, a growing area of monetization growth for publishers is data collaboration. Data collaboration comes from working directly with advertising partners to safely match data in for both audience activation as well as sharing insights about audience traits or purchase behavior. This helps publishers grow their revenue by creating better plans with their advertising partners and offering unique measurement solutions which ultimately leads to bigger commitments and higher CPMs.

‍

Identity Solutions are a key part of Optable

Our Identity Solution is designed to help media businesses adapt to the rapidly changing digital advertising realm. Decision makers need to consider comprehensive identity solutions as a new alternative to third-party cookies to deliver performant targeted campaigns and boost revenue in both direct and programmatic advertising.

Optable offers a comprehensive approach to identity resolution and developing customer ID Graphs, enabling our clients to enhance audience engagement and revenue through better addressability and personalized ad content. By establishing a first-party identity graph and processing second-party data, Optable aims to improve addressability across cookieless environments, enrich audience insights, and unlock new revenue streams through data collaboration. Ask for a demo to learn more.