CUSTOMER SEGMENTATION AND CHURN PREDICTION IN ONLINE RETAIL - PowerPoint PPT Presentation

CUSTOMER SEGMENTATION AND CHURN PREDICTION IN ONLINE RETAIL Authors: Nilay Jha, Dhruv Parekh, Malek Mouhoub, and Varun Makkar Paper: 139 Presented By: Nilay Jha

Outline 1. Introduction 2. Background/Related Concepts 3. The proposed model/architecture 4. Experimentation 5. Conclusion 6. References 2

1) Introduction ● The online retail industry has changed the way customers shop as everything is available online. In order to build a loyal customer base, a company needs to deploy various marketing strategies focused on the diverse nature of its customers. ● A possible solution is to segment customers and make targeted marketing strategies for which historical data of customers is required. ● RFM analysis is a technique that helps in extracting insights from the records and can be used for segmentation of customers as well. However, sometimes RFM(Recency, Frequency, Monetary) analysis alone is not sufficiently insightful. In such situations, it is extended with other variables. ● This paper focuses on improving the process of customer segmentation by extended RFM model, which is named as RFMOC (Recency,Frequency, Monetary, Offer Factor and Category Variance). RFMOC along with one additional variable D is also discussed in the paper for the purpose of churn prediction. 3

2) Background / Related Concepts 4

5

6

3) The proposed model / architecture 7

8

9

10

11

12

13

14

4) Experimentation 15

Results 16

5) Conclusion ● RFMOC performs better than classical RFM when it comes to customer segmentation. RFMOCD performs better than classical RFM when it comes to churn ● prediction. 17

18