Skip to content

Latest commit

 

History

History
33 lines (19 loc) · 1.35 KB

README.md

File metadata and controls

33 lines (19 loc) · 1.35 KB
Raw

E-commerce scoring from Olist Data

Code style: black

How to get the dataset

You can download the dataset and its documentation on Kaggle

INTENT

What is expected for this project is a customer segmentation for Marketing purposes based on the free dataset from Olist company available on Kaggle.

While they also share a secondary dataset about marketing data, we won't use it.

We need to better understand the customers and to provide actionable explanation of the suggested segmentation to the marketing team.

Also, we need to provide an analysis about the stability of the segmentation, to evaluate when they will need to update their segmentation.

While most of their customers do only 1 unique order and the RFM model may be not that efficient, Olist suggest we should categorize customers based on 2 criteria :

  • in terms of order details ;
  • in terms of satisfaction ;

Expected delivery

  • A notebook with exploratory data analysis ;
  • A notebook of tests of different modeling approaches ;
  • A simulation notebook to determine the necessary frequency of updating the segmentation model ;
  • A presentation support to present your work ;

LICENSE

This project is provided under the MIT license.