11-02, 15:20–16:00 (America/New_York), Radio City (Room 6604)
Abstract: In this talk, we will explore the world of Media Mix Modeling (MMM) and examine the application of Bayesian and Frequentist regression methods to derive valuable marketing insights. MMM plays a pivotal role in marketing strategy, enabling businesses to measure marketing effectiveness and optimize budget allocation across various channels. The presentation will offer an overview of both Bayesian and Frequentist approaches, comparing their principles and applications within the MMM context. We will present a real-world case, showcasing how these two models were used to deconstruct sales driven by several factors and generate actionable insights. Finally, we will share reflections and recommendations for marketers and data scientists regarding further exploration and adoption of different techniques in MMM, empowering businesses to make informed decisions and effectively allocate their marketing resources.
In today’s complex marketing landscape, Media Mix Modeling (MMM) is a game-changer. This session delves into MMM’s pivotal role in shaping marketing strategies by comparing Bayesian and Frequentist methodologies.
At its core, MMM offers businesses insight into the performance of different marketing channels, enabling optimal budget allocation.The talk promises a dual-perspective: presenting a comprehensive technical overview while interlacing practical implications and real-world applicability. Beginning with PepsiCo’s specific challenges in sales analytics, attendees will gain an understanding of MMM’s strategic significance. Through practical demonstrations using tools like numpyro, we’ll construct MMM models from both Bayesian and Frequentist perspectives.
Further enriching the session is a PepsiCo case study, illustrating the models’ power in deciphering sales contributions and delivering actionable insights. This case transitions from model-building to its tangible implications in the business realm, making the presentation invaluable for data professionals and marketers alike.
Concluding segments offer reflections on MMM’s expansive potential. By contrasting Bayesian and Frequentist models, attendees will leave equipped with insights to adapt the best-suited approach to their challenges. The session aims to transform participants into MMM mavens, skilled at converting data into business strategies.
This talk caters to a spectrum of professionals: Data Scientists seeking technical rigor, and Marketers eyeing data-driven decision-making. No prerequisites are required, only a zeal for learning.
Agenda Breakdown:
Minutes 1-10: Introduction to MMM and its strategic role.
Minutes 10-20: Building MMM models - contrasting methodologies.
Minutes 20-30: Real-world applications, challenges, and solutions.
Minutes 30-40: Engaging Q&A for knowledge exchange.
No previous knowledge expected
Suri Chen is a Principal Data Scientist at PepsiCo eCommerce. With a background in Operations Research from Columbia University, she possesses expertise in areas including optimization, neural networks, bayesian modeling, consumer clustering, and text mining. As an applied scientist in PepsiCo's cross-functional team, Suri addresses business challenges by harnessing the power of data and machine learning. Beyond her professional role, she is passionate about generative AI research, particularly in the realm of music generation.