Data-driven online service supply chain : a demand-side and supply-side perspective

Purpose: Big data analytics (BDA) and machine learning (ML) can be used to identify the influencing factors of online service supply chains (OSSCs) and can help in the formulation of optimal pricing strategies. This paper analyzes the influencing factors of customer online shopping from the demand-side perspective and formulates optimal pricing strategies from the supply-side perspective. Design/methodology/approach: This paper uses ML and the Stackelberg game approach to discuss OSSC management. ML's feature selection algorithm is used to identify the important influencing factors of 12,330 customers' online shopping intention data using four different classifiers. The Stackelberg game approach is used to analyze the pricing strategies of integrators and suppliers in OSSCs. Findings: First, the feature selection algorithm can improve the efficiency of optimization in big data samples of OSSCs. Second, the level of visualization and the quality of information (page value) will affect the purchase behavior of customers. Finally, the relationship between the optimal pricing and the level of visualization is obtained through the Stackelberg game approach. Practical implications: This paper reveals the phenomenon of “mystery customers,” and the results of this paper can provide insights and suggestions regarding the decision-making behavior of integrators and suppliers in OSSC management. Originality/value: Considering customer behavior intention, this paper uses a data-driven method to explore the influencing factors and pricing strategies of OSSCs. The empirical results enrich the existing OSSC management research, proposing that the level of product visualization and information quality plays an important role in OSSCs.