Beyond Means: A Dynamic Framework for Predicting Customer Satisfaction
This work provides a more accurate method for online reputation systems to predict customer satisfaction, benefiting marketing practitioners and customers, though it is incremental as it builds on existing Gaussian process frameworks.
The paper tackled the problem of predicting customer satisfaction from online ratings by addressing the limitations of standard aggregation methods like the sample mean, which fail to adapt to quality changes over time and ignore review heterogeneity. The result was a Gaussian process model that reduced mean absolute error by 10.2% compared to the sample mean, based on 121,123 Yelp ratings.
Online ratings influence customer decision-making, yet standard aggregation methods, such as the sample mean, fail to adapt to quality changes over time and ignore review heterogeneity (e.g., review sentiment, a review's helpfulness). To address these challenges, we demonstrate the value of using the Gaussian process (GP) framework for rating aggregation. Specifically, we present a tailored GP model that captures the dynamics of ratings over time while additionally accounting for review heterogeneity. Based on 121,123 ratings from Yelp, we compare the predictive power of different rating aggregation methods in predicting future ratings, thereby finding that the GP model is considerably more accurate and reduces the mean absolute error by 10.2% compared to the sample mean. Our findings have important implications for marketing practitioners and customers. By moving beyond means, designers of online reputation systems can display more informative and adaptive aggregated rating scores that are accurate signals of expected customer satisfaction.