DiffuNET: The impact of network structure on diffusion of innovation

Purpose – To provide an explicit model to address the relationships between the structural characteristics of a network and the diffusion of innovations through it. Further, based on the above relationships, this research tries to provide a way to infer diffusion curve parameters (innovation coefficient and imitation coefficient) from network structure (e.g. centralization). Design/methodology/approach – Based on the network and innovation literatures, we develop a model explicitly relating the structural properties of the network to its innovation and imitation potential, and in turn to the observed diffusion parameters (innovation and imitation coefficients). We first employ current theoretical and empirical results to develop postulates linking six key network properties to innovation and imitation outcomes, and then seek to model their effects in an integrative manner. We argue that the innovation and imitation potentials of a network may be increased by strategically re‐designing the underlying network structure. We validated the model by searching the published empirical literature for available published data on network properties and innovation and imitation coefficients. Findings – We validated the model by searching the published empirical literature for available published data on network properties and innovation and imitation coefficients. The results reported from various relevant research papers support our model. Practical implications – This research shows that the innovation and imitation potentials of a network may be increased by strategically re‐designing the underlying network structure; hence, provide guidelines for new product managers to enhance the performance of innovative products by re‐design the underlying network structure. Originality/value – The model developed in this paper is a breaking through result of synthesizing various traditions of diffusion research, ranging from anthropology and economics to marketing which were developed independently. The research explicitly modeled the diffusion process in terms of the underlying network structure of the relevant population allowing managers and researchers to directly link the diffusion parameters to the structural properties of the network. By doing so, it added value by making it possible to infer diffusion potential from directly measurable network properties. Vis‐à‐vis the network diffusion literature in particular, we added value by “unpacking” the diffusion process into innovation and imitation processes that form the building blocks of contagion. Moreover, we developed a holistic structural model of network diffusion which integrates the several network properties that have hitherto been studied separately.