Previous Workshops
2026
Katrijn Gielens, Professor of Marketing at Tilburg School of Economics and Management
Topic: Meet the Editor: Everything you always wanted to know but were afraid to ask
Date and time: June 25, 2026
Abstract
Katrijn's research focuses on the dynamics in digital and brick-and-mortar retailing and relationships between retailers and brand manufacturers. Her work has been published in leading journals such as the Journal of Marketing, Journal of Marketing Research, The International Journal of Research in Marketing (IJRM), Journal of Consumer Research and Marketing Science. She was recently appointed as an Editor of the Journal of Marketing Research, serves as an Associate Editor for the Journal of Marketing, and previously served for many years as Editor of the Journal of Retailing. She has worked extensively with business practitioners, both retailers and brand manufacturers. Her work has received substantial media attention, both nationally and internationally, and she is often consulted by the media to comment on significant retail events.
Timo Mandler, Full Professor of Marketing at NEOMA Business School, France
Topic: AI Across the Academic Workflow: A Pragmatic Look at What's Possible Today
Date and Time: April 9, 2026
Abstract
The capabilities of artificial intelligence for academic research are evolving so rapidly that even well-informed researchers struggle to maintain an accurate picture of what is currently possible — and what is not. Opinions range from dismissive skepticism to uncritical enthusiasm, and both can lead researchers astray. This talk offers a pragmatic, evidence-based assessment of the current state of AI across the quantitative research workflow, with examples tailored to the kinds of tasks marketing scholars routinely face — from exploring and synthesizing large bodies of literature, collecting and processing data, and supporting quantitative analysis to writing and revising manuscripts and disseminating findings. Rather than speaking in generalities, the talk maps specific research tasks to specific AI-based tools and approaches, covering both ready-to-use web-based interfaces and more flexible methods using application programming interfaces. The session also addresses limitations and open questions that come with integrating AI into scholarly practice.
2025
Jan-Benedict Steenkamp, Full Professor of Marketing at UNC Kenan-Flagler
Topic: Empirics First Research
Date and Time: December 16, 2025
Abstract
Empirics First (EF) research is grounded in a real-world marketing phenomenon that involves obtaining and analyzing data and producing valid marketing-relevant insights without necessarily developing or testing theory (Golder et al. 2023). The approach lends itself well to today’s environment, with newly emerging data sources and marketing phenomena that can reveal novel insights and research questions. Empirics first does not mean empirics only. The EF approach serves theory development by starting with data in novel domains, domains untethered to existing theory, or domains with multiple relevant and possibly conflicting theories. Throughout, attention is paid to the existing literature, with an eye toward generating new theory, concepts, constructs, and conceptual frameworks, as well as uncovering empirical regularities and providing guidance to marketing’s stakeholders. The Journal of Marketing has announced a special issue on EF. In this session, Prof. Steenkamp will discuss the key success factors for EF research in general and for the special issue in particular.
Dr. Francisco Villarroel Ordenes, Full Professor of Marketing at the University of Bologna
Topic: Quant Studies in Social Media Marketing: Data, Measurement, Modeling, and Apps
Date and Time: October 23, 2025
Abstract
This workshop shares insights from quantitative research using large-scale social media field data. I will share practical challenges and solutions related to data collection (e.g., web scraping), developing measurements from multimodal content (e.g., text and image analysis with machine learning), and addressing modeling and endogeneity concerns (e.g., robustness checks). The session will also highlight how research findings can be translated into marketing applications, such as prototyping interactive decision-support tools. To illustrate, I will draw on two recent projects: one investigates how the design of text overlays in brand posts shapes consumer engagement, and the other examines how virtual influencers can compose images to increase impact. The first project has a more quantitative focus through the use of vision transformers to classify brand images into dynamic or static, while the second combines quantitative analyses with an experimental approach to find psychological mechanisms. Together, these examples demonstrate how rigorous quantitative approaches can both advance theory and yield actionable insights for social media marketing.
Zhenling Jiang, Assistant Professor of Marketing, The Wharton School, University of Pennsylvania.
Title: Training (and pre-training) neural networks for structural model estimation.
Date and time: June 18, 2025
Abstract
We propose a novel approach to structural econometric estimation by training neural networks to recognize model parameters from data patterns. The neural net estimators are trained on simulated datasets generated under a range of parameter values, with moments of each dataset serving as inputs. Once trained, the estimator delivers fast and accurate parameter estimates. Compared with standard approaches, such as SMLE, the neural net estimator can outperform in both accuracy and computational efficiency.
We further extend the method by pretraining estimators for consumer sequential search models. The estimator is “pretrained” in the sense that the bulk of the computational cost and researcher effort occur during the construction of the estimator. Subsequent applications of the estimator to different datasets require little computational cost or researcher effort. Evaluated across 14 real datasets, the estimator runs in seconds and requires minimal researcher effort. By front-loading the computational cost during training, this method offers a scalable, accessible tool for structural estimation, especially in models that are otherwise computationally intensive.
Christian Hotz-Behofsits, Assistant Professor at WU Vienna
Topic: Natural Affect DEtection (NADE): Using Emojis to Infer Emotions from Text
Date and Time: May 5, 2025
Abstract
Emotions are central to consumer communications, and extracting them from user-generated online content is crucial for marketers, as consumer opinions significantly shape brand perceptions, influence purchase decisions, and provide essential insights for marketing analytics. To leverage vast user-generated data, marketers and researchers require advanced text-to-emotion converters. However, existing tools for fine-grained emotion extraction face several limitations: lexica are constrained by their dictionaries, machine learning models by human-annotated training data, and large language models by insufficient validation. As a result, marketing research still often relies on basic sentiment detection instead of extracting more nuanced emotions from text.
We propose Nade (Natural Affect DEtection) - a novel, text-to-emoji-to-emotion converter that addresses these shortcomings through a two-stage architecture. In Stage I, an embedding-based multi-label classifier predicts 151 emojis, all selected from the Smileys & Emotion group defined by the Unicode standard. These emojis serve as freely available, user-labeled emotion proxies. The classifier was trained on 110 million generic English-language social media posts from Twitter/X. In Stage II, we use gradient boosting trained on 5,975 emotion-intensity-labeled words from the NRC lexicon to map emoji outputs to emotional intensity scores for eight core emotions based on Plutchik's theory. Plutchik's model is widely used in marketing and psychology and offers a structured yet flexible hybrid model that combines the benefits of basic emotions and dimensional emotion theories. While we adopt it as a default, Nade is not restricted to a specific emotional theory: we demonstrate its extensibility by adapting it for Valence-Arousal-Dominance (VAD). Using human raters and state-of-the-art converters as benchmarks, we validate Nade, establish the benefits of exploiting emojis, and showcase several marketing applications across diverse social media platforms. Nade scales efficiently, preserves user privacy through local inference, and is freely available as a web app and R/Python packages - offering a validated and accessible approach to emotion extraction in marketing and beyond.
Klaus Miller, Assistant Professor of Marketing at HEC Paris.
Topic: Who benefits from the data economy? A perspective on the economic value of user tracking for publishers using Augmented Inverse Probability Weighting (AIPW).
Date and Time: January 16, 2025
Abstract
Regulators and browsers increasingly restrict user tracking to protect users’ privacy online. In two large-scale empirical studies, we study the economic implications for publishers relying on selling advertising space to finance their content. In our first study, we draw on 42 million ad impressions from 111 publishers covering EU desktop browsing traffic in 2016. In our second study, we use 218 million ad impressions from 10,526 publishers (i.e., apps) covering EU and US mobile in-app browsing traffic in 2023. The two studies differ in the share of trackable users (Study 1: 85%; Study 2: Apple: 17%, Android: 91%). Still, we find similar average ad impression price decreases (Study 1: 18% and Study 2: 23%) when user tracking is unavailable. More than 90% of the publishers realize lower prices when selling ad impressions for untrackable users. Publishers offering content on sports, cars, lifestyle & shopping, and news & information suffer the most. Premium publishers with high-quality edited content and strong reputations, thematic-focused (niche) publishers, and smaller publishers suffer less from the unavailability of user tracking. In contrast, non-premium publishers with non-edited or user-generated content, thematic-broad (general news) publishers, and larger publishers suffer more. The availability of a user ID generates the highest value for publishers, whereas collecting a user’s browsing history, perceived as intrusive by most users, generates only a small value for publishers. These results affirm that ensuring user privacy online has substantial costs for online publishers, but those costs differ across publishers and the type of collected data. This article offers suggestions to reduce these costs.
