2025-12-10 –, Horace Mann
Measuring the true incremental impact of media spend remains one of the toughest problems in marketing, especially in an era where privacy limits user-level tracking. This talk examines how geo-lift incrementality testing can be utilized to accurately measure the true causal impact of marketing and media channels. Attendees will learn what design decisions matter, how to analyze results, and common pitfalls to avoid when running marketing incrementality tests. The goal is to bring causal inference theory into real-world measurement, enabling practitioners to make informed, data-driven decisions with confidence.
Attribution in marketing is notoriously difficult. Most marketers rely on observational data or heuristics that measure correlation rather than causation, leading to unreliable estimates of channel performance. As privacy regulations limit user-level tracking, the need for experiment-based causal measurement has become more important.
This talk explores geo-lift incrementality testing, a causal inference framework that uses geographic units as the unit of experimentation to measure the true incremental impact of marketing spend. Attendees will learn how to design these experiments from the ground up: selecting control and treatment geographies, ensuring statistical power, accounting for seasonality, and analyzing lift using appropriate statistical models.
The session emphasizes both conceptual foundations and practical implementation using open-source Python tools. A real-world marketing channel spend example will demonstrate how to estimate incremental lift, quantify uncertainty, and interpret findings to guide better budgeting and optimization decisions.
While focused on marketing measurement, these methods are more broadly applicable to any domain where interventions vary across geographic units or groups, such as public policy, healthcare, product rollouts, and promotions.
By the end of this session, attendees will:
- Understand when and why geo-lift incrementality testing is the right causal measurement framework
- Know how to design, analyze, and interpret geo experiments with statistical rigor
- Recognize and avoid common pitfalls that bias incrementality estimates
- Learn how open-source Python tools can be used for scalable causal experimentation
This talk brings together theory and application of causal inference, demonstrating how measuring causal impact can inform real-world decision-making.
Bryce Casavant is a Senior Data Scientist at WHOOP, specializing in experimentation and marketing analytics. He applies causal inference and geo-lift testing to measure the true impact of media and product initiatives, focusing on rigorous statistical results and turning them into actionable business insights that drive smarter business decisions.