Prerelease Buzz Evolution Patterns and New Product Performance

This study examines the dynamics of online buzz over time before product release. Employing functional data analysis, we treat the curve of prerelease buzz evolution trajectory as the unit of analysis and find that the shape of the curve significantly adds power in predicting new product performance compared with using product characteristics and firm advertising alone. Moreover, daily prerelease buzz evolution data enable accurate sales forecasting long before product release, which allows sufficient time for managers to adjust product design and/or marketing strategy. For example, the forecasting accuracy using an early buzz evolution curve ending on the 61st day before product release is not only higher than that using accumulated buzz volume until then but also higher than that using the total volume of all buzz up until product release. Beyond the sales outcome, we find that prerelease buzz is quickly reflected in firm stock returns before product release and reduces the absolute amount of postrelease stock price correction. The model accounts for endogeneity, and the results are robust after controlling for buzz sentiment. We also explore the factors influencing prerelease buzz evolution patterns, thus generating insights into how to manage prerelease buzz dynamics to enhance new product performance.

• 저널Marketing Science · 33(3) · 401–421
• 토픽Innovation Diffusion and Forecasting · Management Science and Operations Research
• DOI10.1287/mksc.2013.0828
• 저자Guiyang Xiong, Sundar G. Bharadwaj