The Effects of Market Network Heterogeneity on Innovation Diffusion: an Agent-Based Modeling Approach
Jonathan D. Bohlmann,Roger J. Calantone,Meng Zhao
DOI: https://doi.org/10.1111/j.1540-5885.2010.00748.x
2010-01-01
Abstract:Innovations usually have an initial impact on very few people. The period of learning or early evaluation precedes the diffusion of the technology into the wider addressed population. More than a transfer, this is best characterized as communication of benefits, costs, and compatibility with earlier technologies and a relative assessment of the new state of the art. Innovation development by an organization or individual creates not just a device (i.e., process or tacit knowledge) but concomitantly a capacity on the part of other organizations or persons to use, adopt, replicate, enhance, or modify the technology, skills, or knowledge for their own purposes. How innovations actually diffuse is to understand the communication of progress, and this framing helps one to design innovations and also design the marketing and testing programs to ready innovations for market and launch them efficiently. Diffusion theory's main focus is on the flow of information within a social system, such as via mass media and word-of-mouth communications. This theory presents often in the form of mathematical models of innovation and imitation. Distinct from classical diffusion models, however, consumers are not all identical in how they connect to others within a market or how they respond to information. We examine the effects of various network structures and relational heterogeneity on innovation diffusion within market networks. Specifically, network topology (the structure of how individuals in the market are connected) and the strength of communication links between innovator and follower market segments (a form of relational heterogeneity) are studied. Several research questions concerning network heterogeneity are addressed with an agent-based modeling approach. The present study's findings are based on simulation results that show important effects of network structure on the diffusion process. The ability to speed diffusion varies significantly according to within-and cross-segment communications within a heterogeneous network structure. The implications of the present approach for new product diffusion are discussed, and future research directions are suggested that may add useful insights into the complex social networks inherent to diffusion. A simple summary is that discovery of significant prime communicator nodes in a network allows innovation development practices to be better calibrated to realistically multiple market segments.