Special Session: Data Analytics Methods for Marketing Strategy Researchers: An Abstract

Abstract

Methods research has been a core part of the marketing discipline for over 50 years. Methods researchers provide tools both for academic researchers to explore marketing phenomenon and for marketing practitioners to better analyze commercial data. There are several reasons for the continuing relevance of methods research in marketing. First, in both the commercial world and in academia, there is an increasing focus on methodology for dealing with the increasingly large and complex data generated by modern business. These data include, but are not limited to, data from internet search, smartphone app/location tracking, customer systems, social media, online reviews, biometric systems, and data-enabled appliances. Work on such data, often comes under the banner of big data analytics and academic researchers require knowledge of new methods to work with these data. Second, methodological courses for business doctoral students increasingly incorporate advanced data analytics methods, with students being trained in widely used programming languages for analytics, such as R and Python. These reasons give a great deal of potential for methods researchers in marketing to “meet in the middle” with applied marketing strategy researchers.

This session is designed to introduce marketing strategy researchers to recent marketing methods and analytics work and to facilitate collaboration between methods researchers and strategy researchers in marketing. This will be done by introducing three different methods. For each method, there will be a non-technical discussion of the method and its assumptions. A short hands-on application of the method will be given, along with references to more detailed resources (e.g., documentation, publications, and tutorials), and a discussion of how best the technique can be employed by strategy researchers.

The three chosen methods cover important areas of marketing modeling and data analytic research, with commonalities in the analysis of consumer brand behavior and brand positioning. The first method allows researchers to build and validate a range of brand equity indices from web-search data. The second method provides an innovative method of brand mapping and positioning analysis, where researchers can analyze the trajectories of brands and how brand competition changes over time. The third method utilizes online reviews and automatic language feature selection to perform a dynamic segmentation of reviewers, which helps characterize different opinions in different segments.