Trendspotting is a popular and important marketing intelligence tool managers can use to identify and track movements in consumer interests and behaviors. Currently, trendspotting is done either qualitatively by trend hunters, who comb through everyday life in search of signs indicating shifts in consumer needs and wants, or quantitatively by analysts, who monitor individual indicators, such as how many times a particular keyword has been searched, blogged, or tweeted online. However, the abundance of real-time marketing intelligence is creating new challenges for the modern marketer; the sheer dimensionality of the available market indicators can be so overwhelming that limited insights can be gleaned by analyzing these data piecemeal. Moreover, the Internet and other electronic media allow consumers to acquire and share information at much faster pace, leading to more frequent and sudden shifts in their needs and wants and resulting in a highly fluid marketing environment that is difficult to manage or adapt to. As a result, modern marketers must constantly monitor a wide range of indicators on consumer interests and behaviors and would benefit greatly from a scientific approach that allows them to distill multitudes of noisy individual signals into a few key common underlying trends.
In this study, the authors demonstrate how trendspotting can be performed quantitatively by systematically uncovering major trend lines hidden behind a large number of observed signals over time. By developing a state-of-the-art structural dynamic factor-analytic model and applying it to multitudes of time series, they are able to distill a large set of noisy individual signals into a few key smoothed latent trend lines that isolate seasonal movements from nonseasonal shifts in consumer interests and behaviors. They demonstrate this novel approach to trendspotting with an application in the U.S. automotive industry using online keyword search data from Google Insights for Search, in which they analyze search volume patterns across 38 major makes of light vehicles over an 81-month period to uncover key common trends in consumer vehicle shopping interest. Their results show that online consumer interest tracking services such as Google Insights for Search can be mined more effectively as a powerful source of marketing intelligence in spotting major trends in the marketplace. For example, the authors show that, during the 81-month period analyzed in this study, 74% of variance in new vehicle sales in the United States can be explained by seven latent trends in consumer online search behavior, which they extracted from the Google data using the structural dynamic factor-analytic model. The proposed method for quantitative trendspotting can also be applied to identifying hidden trendlines from more traditional sources of marketing intelligence, such as customer feedback data collected through repeated cross-sectional surveys, sales data tracked by syndicated service providers, and customer interaction data from firms’ transaction databases.
Rex Yuxing Du is the Marvin Hurley Associate Professor of Marketing in the Bauer College of Business at the University of Houston. He received his PhD in Marketing from Duke University. Professor Du has published in various leading marketing journals, including Journal of Marketing Research, Journal of Marketing, Marketing Science, and Journal of Consumer Research. In 2003, he won the Alden G. Clayton Award for the Marketing Science Institute Dissertation Competition. In 2007, he was a finalist for the Marketing Science Institute/H. Paul Root Award for significant contribution to the advancement of the practice of marketing. In 2009, he received a prestigious nomination as a Marketing Science Institute Young Scholar. In 2011, he won the Best Paper Award at the AMA Advanced Research Technique Forum. Professor Du’s current research interests lie in online consumer interest tracking and trendspotting, interactive marketing, customer relationship management and database marketing, new product diffusion, and sales forecasting.
Wagner A. Kamakura is the Ford Motor Company Professor of Global Marketing in the Fuqua School of Business at Duke University. Professor Kamakura holds a PhD in Marketing from the University of Texas at Austin, a Doctor Honoris Causa from Universidad de Granada, an MS in Industrial Engineering from the University of Sao Paulo (Brazil), an EMBA from the Getulio Vargas Foundation (Brazil), and a BS in Mechanical Engineering from the Institute of Technological Aeronautics (Brazil). He is a coauthor of Market Segmentation: Conceptual and Methodological Foundations, as well as more than 90 articles in academic journals in marketing and other disciplines. He has served as the editor of Journal of Marketing Research, area editor of Marketing Science, and associate editor of Journal of Consumer Research and as a member of the editorial boards of International Journal of Research in Marketing, Journal of Business Research, and Journal of Retailing.Journal of Marketing Research, Volume 49, Number 4, August 2012
View Table of Contents