Bank Customer Segmentation
Feature Engineering and Data Processing
Applying T-SNE and PCA
Clustering Techniques
Refined insights & Model Proposal
A Bank aims to enhance its credit card customer base penetration for the upcoming financial year. The marketing research team has highlighted the potential for market expansion, prompting the marketing department to propose personalized campaigns targeting both new and existing customers. Additionally, market research revealed a negative perception of the bank’s support services, prompting the operations team to seek improvements in service delivery. Consequently, both the head of marketing and the head of delivery have turned to the Data Science team for assistance.
The objective is to segment the existing customer base into distinct groups based on their spending patterns and past interactions with the bank. This segmentation will facilitate targeted marketing campaigns and aid in improving service delivery to enhance customer satisfaction and bank performance.
Through this Project, I have developed skills at :
- Utilized t-SNE: Implemented t-SNE to visualize high-dimensional customer data in a lower-dimensional space, allowing for better exploration and understanding of customer clusters.
- Employed PCA: Applied PCA to reduce the dimensionality of the dataset while preserving most of its variance. This helped in simplifying the dataset and identifying key features driving customer segmentation.
- Optimized Marketing ROI: By segmenting customers and personalizing marketing efforts, the organization achieved higher engagement rates, increased revenue, and improved ROI.