Segmenting Online Shoppers: A Combined Cluster and Logistic Regression Approach for Forecasting Purchase Behavior

Abstract

This paper examines customers’ online shopping behavior and its relationship to purchase decisions. There are two research questions for the paper: (1) How can marketers segment prior customers’ online shopping behavior to develop a more accurate consumer segmentation for forecasting purchase behavior? (2) Which factors influence customers’ online shopping behavior? The analysis used a dataset of 12,330 samples with ten continuous and eight categorical variables. Cluster and logistic regression analyses were conducted on the dataset. Cluster analysis grouped customers by operating system, browser, location, and traffic type, while logistic regression estimated online customers’ purchase intentions. The results show that different factors influence various customer segments. In Group 1, administrative factors, product-related duration, exit rate, and page value significantly affect purchase behavior. In Group 2, product-related factors, exit rate, page value, and special days play an important role. In Group 3, product-related factors, bounce rate, page value, and visitor type are key factors. This paper contributes by using cluster analysis and logistic regression to segment online customer usage behavior and forecast purchase behavior. It evaluates key predictors of online shopping behavior that differ across customer groups, supporting targeted marketing strategies and decision-making.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.