30/04/2013 - 13:00
Dr. Danny Shapira
The Guilford Glazer Faculty of Management
Ben –Gurion University of The Negev
Complex Systems in Marketing and Innovation DiffusionIn this talk we will discuss how complex system analysis can be employed in the study of innovation diffusion and we will focus on two papers: In the first paper entitled “Network Traces on Penetration: Uncovering Degree Distribution from Adoption Data”, done in collaboration with Yaniv Dover and Jacob Goldenberg we show how networks modify the diffusion curve by affecting its symmetry. We demonstrate that a network's degree distribution has significant impact on the contagion properties of the subsequent adoption process, and propose a method for uncovering the degree distribution of the adopter network underlying the dissemination process, based exclusively on limited early-stage penetration data. While previous works have used models to show that a given pattern of adoption can be explained by network effects (although alternative explanations can, in principle, be offered), in this paper we propose and empirically validate a unified network-based growth model that links network structure and penetration patterns. Specifically, using external sources of information, we confirm that each network degree distribution identified by the model matches the actual social network that is underlying to the dissemination process. We also show empirically that the same method can be used to forecast adoption using an estimation of the degree distribution and the diffusion parameters, at an early stage (15%) of the penetration process. We confirm that these forecasts are significantly superior to those of three benchmark models of diffusion. The second part of the talk will be focused on a paper entitled “Zooming In: Self Emergence of Movements in New Product Growth” done in collaboration with Jacob Goldenberg and Oded Lowengart. In this paper we propose an individual-level approach to diffusion and growth models. In our study, we refer to a unit of analysis, which is a single consumer (instead of segments or markets) and use granular sales data (daily) instead of smoothed (e.g., annual) data as is more commonly used in the literature. By analyzing the high volatility of daily data, we show how changes in sales patterns can self-emerge as a direct consequence of the stochastic nature of the process. Our contention is that the fluctuations observed in more granular data are not noise, but rather consist of accurate measurement and contain valuable information. By stepping into the noise-like data and treating it as information, we generated better short-term predictions even at very early stages of the penetration process. Using a Kalman-Filter-based tracker, we demonstrate how movements can be traced and how predictions can be significantly improved. We propose that for such tasks, daily data with high volatility offer more insights than do smoothed annual data.