As retail industry emerges there is increasing motivation for retailers to look for data
or strategies that can help them segment or describe their customers in a succinct, but
informative manner. This work focuses on Customer Segmentation. By integrating
machine learning practices and conventional business understandings, the paths to
understand customer behaviour became more intertwined and answers question like
What segments or groups of customers do we have?.
The tasks performed in this work include customer segmentation, customer classification, RFM Analysis and Market Basket Analysis. The findings are that there are
roughly five or six clusters of customers with each cluster having unique purchasing
traits that define them.
- To cluster customers based on common purchasing behaviors for future operations/marketing projects
- To classify the new customers using the knowledge from previous clustering
- RFM Analysis to understand the customer behaviour
- Market Basket Analysis to know the items buying pattern of the customers
- November has the highest number of sale
- Highest number of customers has arrived on Thursday
- 11AM - 1PM is the prime time for sale
- Customers from UK has made the most orders
- Most income is from UK
- The best set of customers are from Netherlands, Australia, Singapore.
- Most prevailing items are lunch bag, doll, candles etc
Click here to view final report.