A Novel Group Decision Making Model to Compare Online Shopping Platforms

Over the years, E-commerce industry has been witnessing a phenomenal growth, thanks to rapid technological advancement in Industry 4.0. There has been a standout surge in the use of various online shopping platforms (OSP) for daily use. The recent pandemic has accelerated the growth trajectory and made a transformational change in the digital commerce landscape. As a result, there has been a proliferation of OSPs in the competitive domain. It is therefore pertinent to address the questions: How do the customers select their favorite OSP? To what extent the OSPs differ based on consumers’ preferences? The present work addresses these questions by proposing a novel group decision making framework. The ongoing study provides several innovative extensions of multi criteria decision making models like Borda count, criteria importance assessment (CIMAS), modified preference selection index (MPSI), and root assessment method (RAM). In this paper, the researchers provide a novel use of the Borda count method, integrated with CIMAS for determining criteria weights utilizing ranking of the criteria. Further, a novel extension of MPSI and RAM has been made with multiple normalizations. In this paper, the authors demonstrate a rare combination of vector and non-linear normalization using the Heron mean. The present paper derives the final criteria weights by combining Borda count, CIMAS and multi-normalization based MPSI (MNMPSI) using Bayesian logic. The criteria are selected based on Uses and Gratification theory (UGT). The findings reveal that interactive app interface and features (C16), user-friendly interface and search (C13), convenience in shopping (C14), product availability and variety (C12) and discounts and offers (C8) exert significant influence in selecting the OSP. Further, it is observed that Flipkart (A2) and Amazon (A1) are the top performers in the eyes of the users. The stability and reliability of the proposed methodology are examined by conducting a sensitivity analysis and comparing with several other models. The robustness of the proposed methodology and practical relevance of the findings of the present work shall provide notable impetus to the analysts and strategic decision-makers.