The main objective of this project was to analyze a retail business’s customer data to identify patterns and segment customers into distinct clusters. These clusters were used to develop targeted marketing strategies, improve customer retention, and optimize the allocation of resources.

Data:

The dataset used in this project contained customer data with the following attributes:

• CustomerID
• Gender
• Age
• Annual Income (k$)
• Spending Score (1-100)

Methodology:

We employed the K-means clustering algorithm to create distinct customer segments based on their attributes. The optimal number of clusters was determined using the Elbow Method, which identified six clusters as the most suitable representation of the data.

Results:

The six clusters identified were as follows:

Cluster 0: Younger customers with low income and high spending scores

• Age: Mostly younger individuals
• Annual Income: Lower income
• Spending Score: High spending scores

Cluster 1: Middle-aged to older customers with moderate income and moderate spending scores

• Age: Middle-aged to older individuals
• Annual Income: Moderate income
• Spending Score: Moderate spending scores

Cluster 2: Customers of various ages with high income and low spending scores

• Age: Various age groups
• Annual Income: High income
• Spending Score: Low spending scores

Cluster 3: Customers of various ages with moderate income and moderate spending scores

• Age: Various age groups
• Annual Income: Moderate income
• Spending Score: Moderate spending scores

Cluster 4: Customers of various ages with high income and high spending scores

• Age: Various age groups
• Annual Income: High income
• Spending Score: High spending scores

Cluster 5: Younger to middle-aged customers with low income and low spending scores

• Age: Younger to middle-aged individuals
• Annual Income: Lower income
• Spending Score: Low spending scores

Dashboard:

We created a modern, interactive dashboard using Plotly to present the results, which included:

• Distribution of the genders of the clients within the cluster.
• Characteristics of each cluster, including age, income, and spending score.
• A menu to select whether to display information for selected clusters only or all clusters.

Conclusion:

This project provided valuable insights into customer behavior and preferences, which can be used to inform marketing strategies and improve customer retention. By segmenting customers into distinct clusters, the retail business can tailor its offerings and promotions to better meet the needs of its customer base, ultimately leading to increased revenue and customer satisfaction.