Webinar: Integrating Hedonic and Analytic Data for Product Optimization with Bayesian Networks and GenAI
A Case Study from the Fragrance Industry
Webinar Recorded on May 22, 2025
Webinar Recording
Webinar Summary
The Perfume Study Revisited
In this webinar, we revisit one of our most recognized case studies on Probabilistic Structural Equation Modeling (PSEM), Key Driver Analysis, and Product Optimization. This example has been a cornerstone of BayesiaLab courses and seminars and is featured in Chapter 8 of our e-book (opens in a new tab).
The original study was based on a consumer survey on perfumes conducted by a market research agency in France. It illustrates the development of a PSEM using Bayesian networks to identify the key drivers of product liking and to guide product optimization based solely on hedonic evaluations from consumers.
In this updated version, we enrich the case study with two new and distinct data sources to reflect both perceptual and technical aspects of product evaluation:
-
Open-ended responses from consumers, recorded as text data, are first used to estimate structural priors that guide the unsupervised learning of probabilistic relationships among the hedonic dimensions of each product. Subsequently, these textual data are also leveraged to construct a causal network for characterizing overall liking, using the ICI (Independence of Causal Influence) modeling approach.
-
Analytical sensory data from trained assessors, who provide repeated, objective measurements of the sensory attributes of each product. These tests follow strict methodological protocols to ensure reliability and repeatability, yet, by design, exclude hedonic judgment and are conducted with a small panel of experts (n ≈ 10).
This dual data approach presents a classical challenge in sensory science: analytic and hedonic tests are grounded in fundamentally different epistemologies and require different types of assessors. While trained panels provide the precision necessary for technical product optimization, they do not represent consumer perception. Conversely, consumer data captures market-relevant hedonic reactions but lacks the objectivity and granularity needed for targeted reformulation.
In this webinar, we demonstrate how to construct a Bayesian Network that integrates these two complementary perspectives, merging the objective insights of trained panels with the real-world preferences of consumers, to bridge the gap between sensory science and market performance.
About the Presenters
Dr. Lionel Jouffe
Dr. Lionel Jouffe is co-founder and CEO of France-based Bayesia S.A.S. Lionel holds a Ph.D. in Computer Science from the University of Rennes and has worked in Artificial Intelligence since the early 1990s. While working as a Professor/Researcher at ESIEA, Lionel started exploring the potential of Bayesian networks.
After co-founding Bayesia in 2001, he and his team have been working full-time on the development of BayesiaLab. Since then, BayesiaLab has emerged as the leading software package for knowledge discovery, data mining, and knowledge modeling using Bayesian networks. It enjoys broad acceptance in academic communities, business, and industry.
Stefan Conrady
Stefan Conrady has over 20 years of experience in decision analysis, analytics, market research, and product strategy, having worked with Mercedes-Benz, BMW Group, Rolls-Royce Motor Cars, and Nissan across North America, Europe, and Asia.
As Managing Partner of Bayesia USA and Bayesia Singapore, he is widely recognized as a thought leader in applying Bayesian networks to research, analytics, and decision-making. Together with his business partner, Dr. Lionel Jouffe, he co-authored Bayesian Networks & BayesiaLab — A Practical Introduction for Researchers, an influential resource now widely cited in academic literature.
With their deep expertise in Bayesian networks for Key Driver Analysis and Optimization, Stefan and Lionel are highly sought-after consultants, advising global leaders such as Procter & Gamble, Coca-Cola, UnitedHealth Group, L’Oréal, the World Bank, and many of the world’s largest market research firms.
Similar Studies and Applications
While this case study focuses on sensory analysis, the same approach applies to many other domains. In the past, we have shown similar workflows in the context of consumer or employee surveys with the objective of improving overall satisfaction.