Estimation of the Probability of Success Based on Communication History

The paper presents a study of how the detailed history of communication between a sales agent and a customer can be used to estimate the probability of success (sale). It utilizes the data obtained from a Minnesota based company that uses telesales in a business-to-business environment. Extensive telesales database was investigated and principal variables influencing the sale event were determined. The number of decision-maker contacts X is analyzed as an explanatory variable. A comparison is made between logistic regression technique and discriminant analysis based on negative binomial or zero-modified Poisson distribution of the number of contacts. There seems to be a good agreement between the results obtained by both methods. Discriminant analysis as demonstrated by Efron (1975) can be more efficient if some model assumptions are made regarding the distribution of X . However, logistic regression may be used in order to validate the choice in a certain class of models. A. Introduction and Methods A