Extracting Dimensions of Consumer Satisfaction with Quality from Online Chatter: Strategic Brand Analysis of Big Data Using Latent Dirichlet Allocation
61 Pages Posted: 15 Mar 2014 Last revised: 20 Mar 2014
Date Written: March 13, 2014
Online chatter or User-Generated Content (UGC) constitutes an excellent emerging source for identifying dimensions of quality at a very high temporal frequency. This study proposes a unified framework for extracting the latent dimensions of consumer satisfaction with quality and ascertaining the valence, labels, validity, importance, dynamics, and heterogeneity of those dimensions using unsupervised Latent Dirichlet Allocation (LDA). Our sample of UGC consists of rich data on product reviews across fifteen firms in five markets, over four years. The results suggest that a few dimensions are enough to capture quality. These dimensions have good face and external validity. Dynamic analysis enables tracking the importance of dimensions over time. It also allows for dynamic mapping of competitive brand positions on those dimensions over time. For vertically differentiated markets such as cell phones and computers, objective dimensions dominate and are similar across markets, heterogeneity is low across dimensions, and stability is high across time. For horizontally differentiated markets, such as shoes and toys, subjective dimensions dominate but vary across markets, heterogeneity is high across dimensions, while stability is low over time.
Keywords: Consumer Satisfaction, Quality, Dimensions, Brand Mapping, Big Data, Latent Dirichlet Allocation, Online Chatter, User Generated Content
Suggested Citation: Suggested Citation