Advanced TV Targeting: Deterministic v. Probabilistic Data Providers

October 15, 2019
Advanced TV Targeting

The rise of streaming services has provided consumers with infinite ways to access their favorite video content, while also opening new inventory to advertisers. While ad spots on linear TV were once part of a “spray and pray” advertising approach, the new wave of Advanced TV – encompassing OTT/CTV and Addressable Television – leverages data, science, and software to enable more granular audience segmentation and targeting.

According to eMarketer, the number of US subscription OTT service users will reach 197.7 million by 2020. Yet while OTT currently accounts for nearly 30% of TV viewing, ad-supported services only capture 3% of television ad budgets. Considering that advertisers are still spending up to $70 billion on TV advertising annually, Internet-connected device providers and other streaming players are racing to capitalize on new opportunities in the video advertising space.

The Advanced TV Challenge: Disconnected Ecosystem

Advanced TV enables advertisers to target and serve ads to specific households and subscribers, rather than broadcasting the same messaging to all households. Yet as consumers access video content across mobile, tablet and OTT TV platforms, it becomes increasingly difficult for advertisers to track and measure user activity throughout the customer journey. In order to validate reach and performance across key audience segments, advertisers are looking beyond IP-matching to device-level data.

Probabilistic vs. Deterministic Data Providers

Identity resolution is key to understanding how individuals access campaign messaging across different channels and devices. Whether probabilistic or deterministic, device graph data helps advertisers understand which devices belong to the same household or individual, independent of IP address. Deterministic data providers use a consistent identifier such as a phone number or email address to match a device to an individual and target the same user across multiple screens. When coupled with TV viewing data, deterministic data can further elevate targeting precision by more closely reflecting a brand’s most valuable consumers.

Probabilistic approaches draw from thousands of anonymous data points to identify statistically significant correlations between devices based on usage. For example, if a smart TV, laptop, and tablet connect to the same Wi-Fi network at the same time and in the same places every day, then it can be inferred that all three devices belong to a specific person. Whereas deterministic data providers can authenticate users across devices, probabilistic approaches are based on modeled and/or inferred data, offering expanded reach but less accuracy. While device graph technology for TV measurement is not yet without its flaws, deterministic data can offer advertisers a good foundation for executing a people-based marketing approach on Advanced TV.

The Fluent Advantage

Fluent offers segments comprised of deterministic data, enabling true addressable TV. Consumer insights are sourced exclusively from Fluent’s owned and operated sites. Fluent’s 100% first-party, self-declared data can be used to identify the programs, networks, and dayparts that index against a target audience, helping advertisers achieve household-level targeting among cable or satellite subscribers. Beyond traditional TV, Fluent also enables advertisers to target over 50 million cord-cutters through Connected TV and OTT streaming services across devices. To learn more about Fluent’s solution for targeting television audiences at scale, connect with us here.