Electronic word of mouth analysis for new product positioning evaluation
Journal
Electronic Commerce Research and Applications
Date Issued
July 1, 2020
Author(s)
DOI
10.1016/j.elerap.2020.100986
Abstract
People increasingly choose to express themselves online through electronic word of mouth (eWOM), generating large amounts of data, making eWOM a valuable source of information through big data analytics. This enables organizations to gain insights directly from customers’ opinions for better decision making. This work presents a new methodology for evaluating an organisation's product-positioning strategy through eWOM analytics. A product's mispositioning has significant negative effects and there is strong interest in identifying ways to avoid it. Current methods that utilize eWOM for product positioning evaluation mostly use post-product release reviews and do not statistically evaluate the effect of time on the product positioning; nor do they provide any means to diagnose the cause of mispositioning. The temporal aspect of positioning, however, provides valuable insights into which product features are more time-invariant and accordingly makes it possible to plan for product redesign or repositioning to maximize profitability. A case study is presented in the context of smartphones using design science research, utilizing Twitter data regarding the release of a new product, collected using a custom Android application. The research questions addressed in this paper are: (1) How do consumers’ preferences change over time with regards to the product's positioning? (2) Which product features positively influence product positioning and which negatively? To answer these questions, we compared the product-positioning strategy and consumers’ opinions before and after the release of a new product to identify possible discrepancies between expected and actual positioning of the product. This work constitutes a methodological contribution with demonstrated implications for new product positioning strategy evaluation using tweet analysis.