Analysis of Customer Journeys Using Prototype Detection and Counterfactual Explanations for Sequential Data
Keita Kinjo
TL;DR
This work tackles quantitative analysis of customer journeys by introducing a three-step, distance-based framework for sequential data: (i) a stage-weighted Levenshtein distance to measure similarity and extract prototypes via k-medoids, (ii) 2D visualization of sequences with MDS-like embedding, and (iii) purchase prediction using k-NN plus counterfactual explanations derived from existing sequences to identify actionable path alterations. The approach is validated on survey data from cosmetics purchases, demonstrating identifiable representative journeys and showing how small changes in touchpoints can flip purchase outcomes. The results offer interpretable insights for marketing strategy, enabling targeted interventions and realistic scenario explorations without generating synthetic data. Overall, the methodology provides a practical tool for understanding and guiding consumer journeys through quantitatively grounded, explainable sequence analysis.
Abstract
Recently, the proliferation of omni-channel platforms has attracted interest in customer journeys, particularly regarding their role in developing marketing strategies. However, few efforts have been taken to quantitatively study or comprehensively analyze them owing to the sequential nature of their data and the complexity involved in analysis. In this study, we propose a novel approach comprising three steps for analyzing customer journeys. First, the distance between sequential data is defined and used to identify and visualize representative sequences. Second, the likelihood of purchase is predicted based on this distance. Third, if a sequence suggests no purchase, counterfactual sequences are recommended to increase the probability of a purchase using a proposed method, which extracts counterfactual explanations for sequential data. A survey was conducted, and the data were analyzed; the results revealed that typical sequences could be extracted, and the parts of those sequences important for purchase could be detected. We believe that the proposed approach can support improvements in various marketing activities.