The study utilized a customer behavior dataset from a grocery store’s database to conduct a clustering analysis aimed at segmenting customers based on their demographics, product spending, and engagement patterns. The dataset comprised 2,240 samples with 29 attributes. After preprocessing steps like handling missing values, encoding categorical variables, and standardizing data, Principal Component Analysis (PCA) reduced the dimensionality of the data. Agglomerative Clustering was applied, and the optimal number of clusters was determined using the Elbow Method, resulting in four distinct customer segments. Cluster 1, consisting of high-income, high-spending customers, accounted for 18.7% of the population and was identified as the most valuable segment. Cluster 3, with low-income but high-spending customers, indicated financial risk, suggesting the need for credit monitoring. Clusters 0 and 2, which made up 63% of the population, represent a core market with opportunities for targeted marketing. The customer profiles revealed differences in family structure, income, and life stages across the clusters. Tailored strategies were recommended: exclusive loyalty programs for Cluster 1, flexible payment plans for Cluster 3, and family-oriented services for Clusters 0 and 2. By adopting these strategies, the grocery store can enhance customer satisfaction, improve resource allocation, and optimize market competitiveness.
Yiyang Song (Thu,) studied this question.