Marketing insights from text analysis

Authors: Jonah Berger, Grant Packard, Reihane Boghrati, Ming Hsu, Ashlee Humphreys, Andrea Luangrath, Sarah Moore, Gideon Nave, Christopher Olivola, Matthew Rocklage

Abstract

Language is an integral part of marketing. Consumers share word of mouth, salespeople pitch services, and advertisements try to persuade. Further, small diferences in wording can have a big impact. But while it is clear that language is both frequent and important, how can we extract insight from this new form of data? This paper provides an introduction to the main approaches to automated textual analysis and how researchers can use them to extract marketing insight. We provide a brief summary of dictionaries, topic modeling, and embeddings, some examples of how each approach can be used, and some advantages and limitations inherent to each method. Further, we outline how these approaches can be used both in empirical analysis of feld data as well as experiments. Finally, an appendix provides links to relevant tools and readings to help interested readers learn more. By introducing more researchers to these valuable and accessible tools, we hope to encourage their adoption in a wide variety of areas of research.

Keywords: Natural language processing, Automated textual analysis, Language