Customer Journey Mapping Using Stochastic Models

Abstract

A customer journey map is a visual representation of the process that a person goes through when interacting with a product or a service and it is often related to human-centered design. The process of which customer journey maps are built is referred to as customer journey mapping (CJM) and traditionally this process includes techniques such as observations, interviews, and surveys. However, the emergence of new data collection techniques such as interactive mobile applications has made richer data available for service designers. This emerging data availability poses both challenges and opportunities for CJM. In this paper, we propose an innovative stochastic-based method to tackle these challenges while preserving the advantages of traditional CJM. Specifically, the proposed method models user-generated customer experiences as Markov chains and amalgamate the large quantities of experiences into a small number of customer journey maps which are easier to analyze for service designers. This method is based on a clustering algorithm that achieves the maximum posterior likelihood post-clustering. By employing our method on collected data, service designers can discover key components in customers’ experiences in addition to some edge cases. We will demonstrate the effectiveness of this method using sample data from the 2017 National Household Travel Survey (NHTS).