What Drives Users to Recommend Mobile Fitness Apps? A Three-Stage Analysis Using PLS-SEM, Machine Learning, and fsQCA

In response to the growing demand for studies into recommendation behaviour for mobile fitness apps, this study identified key drivers of users' recommendation intentions through a comprehensive three-stage analysis incorporating PLS-SEM, machine learning, and fsQCA. The PLS-SEM analysis revealed that perceived task facilitation, experiential gratification, and fitness stratification enhanced flow state, while perceived operational facilitation did not. The findings also showed that flow state positively influenced perceived fitness outcomes and recommendation intentions. Machine learning classifiers validated these relationships and highlighted the mediating role of perceived fitness outcomes. The final fsQCA stage identified nine configurations driving recommendation intentions, highlighting the need for app features tailored to user preferences. The findings are important for marketers, app developers, and policymakers. Marketers can design targeted campaigns based on user preferences, while app developers should prioritize intuitive, enjoyable user experiences. Policymakers can use insights to promote standards for user engagement. The study also contributes methodologically by integrating PLS-SEM, machine learning, and fsQCA.