Optimizing Marketing Strategies by Segmenting Customer Purchasing Behavior Using Web Browsing Data

Businesses commonly use RFM analysis to segment their customers into various groups based on purchasing behavior.  This segmentation is calculated by assigning scores to each of the three dimensions, namely, recency, frequency, and monetary, to identify customer segments that may require different marketing strategies or service approaches.  While effective in understanding historical data, the RFM model is limited in its ability to predict future behaviors and adapt to individual customer journeys influenced by digital interactions.  This limitation underscores the need for a new approach. 

This capstone project analyzed web browsing data from a promotional product distributor’s e-commerce site to find key indicators influencing visitors’ purchasing decisions.  The analysis used a double-layer clustering algorithm to identify behavioral purchasing patterns and segment customer data into distinct clusters.  This methodology successfully identified six types of personas: High-Spender Customer, Medium-Spender Customer, Low-Spender Customer, High Purchase Intent Visitor, Low Purchase Intent Visitor, and No Purchase Intent Visitor.  These classifications further revealed web visitor profiles that captured customers’ diverse needs, preferences, and motivations within the promotional product market.  The personas reflected various shopper types, including Conscious Shoppers, Impulse Shoppers, Cautious Shoppers, Exploratory Browsers, and Casual Window Shoppers, each exhibiting unique purchasing behaviors.  This nuanced understanding allowed the marketing team to develop personalized marketing strategies for improving customer acquisition and retention rates, thereby increasing marketing efficiency and effectiveness.