Reviewing 3G “Optimism”: Services, Segments and Stages of Reality

This paper starts by presenting a framework to compare approaches used to forecast the demand for third-generation (3G or UMTS) mobile telecommunications. This analysis envisions the models as forming a spectrum according to how finely they divide the potential market. The division may be in terms of particular services offered, devices used, and/or demographics. The spectrum ranges from the polar extremes of a single, undifferentiated market to models with ten or more services. Given the tradeoffs between massive data needs, model complexity, and the difficulties in aggregating demands across services and segments, it appears that there is an advantage to having a model with a limited number of services along with an explicit approach to the aggregation problem. We then discuss how to incorporate mobile network coverage decisions and demographic changes into a diffusion model. In the final sections we present a first look at how successful one set of 3G demand projections has been and concluding remarks.