Imitation analysis Early prediction of the market demand for major innovations David J. Langley and Nico Pals TNO, Netherlands Organisation for Applied Scientific Research, Groningen, The Netherlands J. Roland Ortt Delft University of Technology, Delft, The Netherlands, and Tammo H.A. Bijmolt Faculty of Economics, University of Groningen, Groningen, The Netherlands Abstract Purpose – The purpose of this paper is to describe a method of estimating the likelihood that a person with particular characteristics will imitate a particular new behaviour (i.e. the use of an innovation). This estimation can be used to provide a new form of forecast for the likely market demand for an innovation. Design/methodology/approach – This method, termed imitation analysis, is based on imitation theories from the behavioural sciences and is applied in two recent case studies in The Netherlands: broadcast TV on mobile phones and a mobile friend-network service. Findings – These cases illustrate how: the market segments with the highest potential can be identified; marketing communication can be focused on specific issues important for each segment (e.g. based on the highest imitation potential); product design can be improved (by highlighting the characteristics with the most room for improving the imitation potential); and market demand can be modelled (the overall chance of imitation occurring). Practical implications – Management implications for the two services, as well as the usefulness of imitation analysis in forecasting studies, are discussed. Originality/value – The paper expands on original work published in this journal in 2005, showing the value of the approach in real-world settings. Keywords Consumer behaviour, Forecasting, Product development, Telecommunications, Supply, demand and exchange Paper type Research paper 1. Introduction The challenge of how to predict the market demand for an innovation prior to its introduction in the market, i.e. when there is little or no data available, is one of the main challenges facing the forecasting community (Meade and Islam, 2006). This key topic has been studied from different disciplinary angles. Economists and econometrists have studied the market conditions that foster innovation and how to measure them (Schumpeter, 1942; Mansfield, 1968; Kamien and Schwartz, 1975; Allen and Fildes, 2001). Sociologists and psychologists have tried to pinpoint the unique characteristics of the innovators and thereby predict adoption (Engel et al., 1990; Foxall and Goldsmith, 1994; Rogers, 2003). Consumer researchers measured the reaction of potential customers to product concepts and thereby tried to estimate the future The current issue and full text archive of this journal is available at www.emeraldinsight.com/1460-1060.htm Imitation analysis 5 European Journal of Innovation Management Vol. 12 No. 1, 2009 pp. 5-24 q Emerald Group Publishing Limited 1460-1060 DOI 10.1108/14601060910928157