The influence of shopping path length on sales growth and its variance

We study the influence of shopping path length on the number of items purchased. It is a well known fact in in-store shopper marketing that the longer the travel distance, the more items purchased. We analyze a shopping path data of 556 shoppers collected at a retail store in an urban are in Tokyo, Japan. We observed the fact the longer the travel distance, the greater the variance of the number of items purchased as well as its average. To explain this fact, we proposed a probability model of sequential purchase decisions using the Bernoulli process. We can estimate the number of items purchased as the binomial distribution given the shopping path. In the case study, we show how this assumption can be justified by the data from a real store.