Paper Type
ERF
Abstract
The video game industry's shift toward live service models raises a critical question: does sustained engagement reflect sustained satisfaction? This study documents and quantifies an engagement-satisfaction paradox using 9,052,077 Steam reviews across 209 games. Employing transformer-based sentiment scoring (RoBERTa), we find that players accumulate substantially more playtime in live service games yet express consistently lower and more steeply declining sentiment as engagement intensifies. Three nested models confirm the paradox: OLS with game fixed effects establishes the core effect, user fixed effects identification confirms it is a within-person phenomenon rather than a reviewer selection artifact, and quantile regression shows it concentrates in the lower tail of the satisfaction distribution — consistent with a compelled-play mechanism rooted in obsessive rather than harmonious engagement. Findings have implications for IS research on platform dynamics and for platform governance and consumer welfare policy.
Paper Number
1186
Recommended Citation
Naeimaeimousavi, Amir and Gupta, Ashish, "Understanding the Paradox in Live Service Games: Evidence from User Reviews" (2026). AMCIS 2026 Proceedings. 2.
https://aisel.aisnet.org/amcis2026/sig_dsa/sig_dsa/2
Understanding the Paradox in Live Service Games: Evidence from User Reviews
The video game industry's shift toward live service models raises a critical question: does sustained engagement reflect sustained satisfaction? This study documents and quantifies an engagement-satisfaction paradox using 9,052,077 Steam reviews across 209 games. Employing transformer-based sentiment scoring (RoBERTa), we find that players accumulate substantially more playtime in live service games yet express consistently lower and more steeply declining sentiment as engagement intensifies. Three nested models confirm the paradox: OLS with game fixed effects establishes the core effect, user fixed effects identification confirms it is a within-person phenomenon rather than a reviewer selection artifact, and quantile regression shows it concentrates in the lower tail of the satisfaction distribution — consistent with a compelled-play mechanism rooted in obsessive rather than harmonious engagement. Findings have implications for IS research on platform dynamics and for platform governance and consumer welfare policy.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIG DSA