From Density to Destiny : Using Spatial Analysis for Early Prediction of New Product Success

Much of the research on new product introductions to date has focused on improving the initial go/no-go decision in order to reduce the probability of post-launch failure. However, one of the main problems associated with early-period assessments is the lack of data to enable further prediction. The few periods of aggregate sales data available to marketers are not enough on which to base reliable predictions. We show that managers can use spatial distribution of sales data to obtain predictive assessment of the success of a new product shortly after launch time. The rationale of our approach stems from diffusion theory. Specifically, we expect that internal influence from previous adopters, i.e., word-of-mouth and imitation, will play a significant role in the success of an innovation. Given initial external marketing efforts by the providers of the new product such as advertising and public relations, early potential adopters will try the innovation first. In order for a product to be a successful, word-of-mouth and imitation effects must take place. Because word-of-mouth spread is often associated with some level of geographical proximity between the parties involved, one can expect that 'clusters' of adopters will begin to form. Alternatively, if the market general reaction is reluctance to adopt the new product, word-of-mouth effect is expected to be significantly smaller, leading to a more sporadic pattern of sales. The small number of users who try the product will be a result in this case of external efforts, leading to a more uniform geographical distribution of adopters. Eventually, such a product will be declared a failure. If word-of-mouth is imperative for successful introductions, and if the formation of clusters can be analyzed to represent its strength, we can define a new discriminating tool between successes and failures. Accordingly, we can examine how 'far' the process is from a uniform geographical distribution. The product whose distribution is further away from a uniform distribution will have a higher likelihood of beginning a 'contagion process' and therefore being a success. To analyze this phenomenon we propose cross-entropy analysis, a method widely used in areas such as physics and information studies to measure the distance between two distribution functions. We apply our model to both simulated and real-life data. We find that cross-entropy has the unique ability to predict success in the beginning of the process, which makes it appealing in marketing activity in general and especially for launches of new products