A longitudinal exploration of the relations between electronic word-of-mouth indicators and firms’ profitability: Findings from the banking industry

Chuanyi Tang, Matthias R. Mehl, Mary Ann Eastlick, Wu He, Noel A. Card

Research output: Contribution to journalArticle

20 Scopus citations

Abstract

Prior research on electronic word-of-mouth (eWOM) has focused on the predictive utility of star ratings. Extending these studies conceptually and methodologically, this paper employed Automatic Text Analysis to investigate the predictive utility of evaluative textual information contained in online reviews. Based on a real-world dataset that matched eWOM with annual financial performance of 68 banks over an eight-year period, this study tested patterns of the bi-directional relations between eWOM indicators and banks’ profitability over time. Results showed that both star ratings and consumers’ verbalized emotions in eWOM significantly predicted increases in firms’ future profitability, which is measured by Return on Assets. Star ratings emerged as a consistent predictor, and their effects lasted for at least two years. Expressed anger predicted lower profitability in the following year and explained additional variance beyond the star ratings. Finally, higher firm profitability was prospectively related to higher star ratings and more verbalized positive feelings in next year's eWOM.

Original languageEnglish (US)
Pages (from-to)1124-1132
Number of pages9
JournalInternational Journal of Information Management
Volume36
Issue number6
DOIs
StatePublished - Dec 1 2016

Keywords

  • Electronic word-of-mouth
  • Profitability
  • Sentiment analysis
  • Social media
  • Text analysis

ASJC Scopus subject areas

  • Information Systems
  • Computer Networks and Communications
  • Library and Information Sciences

Fingerprint Dive into the research topics of 'A longitudinal exploration of the relations between electronic word-of-mouth indicators and firms’ profitability: Findings from the banking industry'. Together they form a unique fingerprint.

  • Cite this