Beyond Means: A Dynamic Framework for Predicting Customer Satisfaction

TL;DR

This paper introduces a Gaussian process-based dynamic framework for aggregating online ratings, improving prediction accuracy by capturing temporal changes and review heterogeneity, thus providing more reliable customer satisfaction signals.

Contribution

The paper develops a tailored Gaussian process model that accounts for rating dynamics and review heterogeneity, outperforming traditional mean-based aggregation methods.

Findings

GP model reduces mean absolute error by 10.2%

GP outperforms sample mean in predicting future ratings

Enhanced rating aggregation improves online reputation systems

Abstract

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.