Newton Research has launched Unlimited Analytics, an agentic AI layer designed to automate ad campaign lifecycles from planning through measurement using causal modeling. Horizon Media has integrated this technology into its Blu marketing platform, reporting improved insights into CTV audience segments and campaign performance.
The launch addresses the industry's shift toward 'agentic' workflows, where AI moves beyond generating text to executing complex data operations and media buying decisions. By automating the 80% of time analysts traditionally spend on data preparation, Newton Research is enabling agencies to focus on high-level strategy and client relations. For the streaming ecosystem, this provides a more scientific alternative to traditional correlations, allowing buyers to verify the incremental impact of CTV ads against broader digital spend. As privacy regulations continue to restrict user-level tracking, look for more platforms to adopt similar causal-based, privacy-safe forecasting models to prove media value.
The rollout follows a period of aggressive agentic AI development within major holding companies and independent agencies. In June 2026, per PR Newswire, Horizon Media expanded its Blu platform with agentic buying capabilities through partnerships with Databricks and media giants including Disney, NBCUniversal, and TikTok. This infrastructure allows agents to make real-time decisions across publishers while maintaining human oversight. This shift toward open, interoperable AI is part of a broader competitive trend; for instance, Horizon and Havas launched the Horizon Global joint venture in late 2025 to create an AI-native agency network designed for rapid speed-to-insight.
Technically, the move toward causal modeling reflects a growing industry mandate for 'triangulated measurement.' Per reports from Measured in June 2026, the cancellation of Google’s Privacy Sandbox cookie replacement and persistent signal loss have forced brands to adopt incrementality testing as a 'causal ground truth.' Industry benchmarks from Newton Research itself suggest its agentic layer is over 300% more accurate than standard large language models (LLMs) for marketing-specific tasks because its agents are pre-trained on marketing science blueprints rather than generic data. This accuracy is increasingly critical in the FAST and CTV sectors, where Omdia and other analysts noted in mid-2026 that content acquisition and ad-tier performance are being optimized through similar machine-learning-driven consumption metrics to manage fixed budget limits.
Add StreamingMeme as a preferred source on Google to see more of our streaming news at the top of your Search results.
Add as preferred source