heisenberg1804 / customer-retail-data-analysis-and-segmentation

Performed an extensive comprehensive exploratory data analysis on Customer Retail Dataset and did kmeans clustering to create customer categories to target various customer groups.

Customer-Retail-Data-Analysis

Conclusion of my EDA: • Number of orders cancelled: 7901/44876 (17.61%)

• 43% of the total purchases have a total price in between 200 and 500 currency and around 60% of the total orders have a total cart price more than 200 currency.

• The company receives high number of orders and highest Total_price from UK as compared to other countries, this means that the company must be UK based.

• There is seasonality in the orders as the sales increase from September till end of Early December. Hence this period is the busiest time for the company maybe because of End of summer Sale Period.

• The number of orders received by the company tends to increase from Monday to Thursday and decrease afterward.

• The company receives the highest number of orders at 12:00pm, this must be because of the lunch hours free time.

• The company gave away maximum no of free items on the month of November 2011, they also gave out 3-6 FREE items to customers each month (Except in June 2011) on average.

Potential Analysis:

1. This dataset also requires a deeper analysis of the cancelled orders.

2. The customers can be segmented according to the Description of the transactions. The keywords can be extracted from the description and the number of customers can be distributed among them. Clusters of keywords could be created and customers could be categorised according to the clusters.

3. Due to the high dimensionality of the data frame a Principal Component Analysis would be required followed by the K means clustering technique to categorise the data.