The Contribution of Unstructured Data and Statistical Learning Models to Marketing Management

Abstract

The capacity to obtain market insights is a strategic need for companies to remain competitive. Despite this and the massive volume of data generated by consumers every second, companies rarely have the culture of making marketing decisions based on data and, when they do, rarely use consumer data widely available online, specially on social networks. One reason is that these data (e.g. texts) tend to be “dirty”, disorganized and bulky, a so-called unstructured data. Despite the complexity involved in extracting informational value from this data, companies can gain insights that can improve decision making and result in greater competitive performance. The purpose of this article is to discuss the benefits of new types of data that have become more abundant and accessible in Web 3.0, as well as new methods of analysis, particularly learning methods. For this, an extensive literature review was carried out and a topic modeling was conducted to get an overview of the data and methods. At the end, the article suggests six main marketing challenges that unstructured data analytics can contribute, improving companies’ competitiveness.