When negative customer engagement turns positive: A four-stage simulation model of engagement evolution in online brand communities
Year: 2026
Type: article
Abstract: This study investigates whether and when service recovery can convert negative customer engagement (NCE) into positive customer engagement (PCE) in online brand communities, and how early value conditions shape later engagement trajectories. Using Xiaomi Community as the focal context, we develop a four-stage threshold-based simulation framework in which consumer utility is treated as a latent evaluation antecedent rather than customer engagement itself. An explicit non-engagement (NE) interval is positioned between the negative and positive thresholds so that moderately evaluated users are not forced into PCE or NCE. Price, product quality, service quality, and personal needs met are retained as the focal antecedents. Core scenarios are repeated 1,000 times with recorded random seeds and 95% simulation uncertainty intervals, and robustness is examined across conversion rates, thresholds, initial conditions, coefficient distributions, sign-allocation rules, and simulation horizons. Price and product quality show strong and opposite effects on the PCE-NCE balance, and a direct price-by-quality analysis indicates that higher quality attenuates the adverse effect of higher price. Service-quality effects remain under both Stage-4-only and continuous-service architectures, but the results do not support a general claim that later-stage service investment is superior to earlier service. Targeted recovery applied to NCE users produces a stronger and more stable NCE-to-PCE repair effect. The results clarify the assumptions, robustness, and managerial boundary conditions of the stage-based model. Because coefficient sign allocation remains a major structural uncertainty, the reported quantities are interpreted as scenario outputs and robust directional comparisons rather than forecasts of actual PCE, NCE, or NE population shares.
Source: PLoS ONE
Authors: Luning Zang, Yuying Liu, Linqiang Wang
Language: English
Subfield: Marketing
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $2,382
OpenAlex ID: W7225805913