This repo is a code demo that implements a custom Customer Retention Analysis class with a number of helpful methods to generate customer churn insights frequently used for marketing analytics to understand the growth and change of an organisation's customer base (new vs retained vs lost)
This repo is accompanied by a Medium blog article "Analyzing Customer Retention Via Cohort Analysis" available at https://medium.com/p/1f381748e555
The code is implemented with three different datasets to demonstrate the ease at which analysis can be run once the data is cleansed and loaded:
- The Sprocket Central Dataset is one year's worth of customer transaction records for a fictional company that makes bicycles and other associated products https://www.kaggle.com/datasets/archit9406/customer-transaction-dataset
- The Online Retail II Dataset which is online retail transaction data set of two years from the year 2009-2010 and 2010-2011 for a UK based e-commerce company https://archive.ics.uci.edu/dataset/502/online+retail+ii
- Transaction Data which is another item purchased transactions dataset with a year's worth of data https://www.kaggle.com/datasets/vipin20/transaction-data
Consider checking out my other repositories too ! :
- https://github.com/ZhijingEu/RFM_Analysis_KMeans - This a simple RFM Analysis Using K Means Clustering On A Publicly Available Brazilian e Commerce Dataset on Kaggle
- https://github.com/ZhijingEu/Customer_Lifetime_Values_BTYD_Modelling_PyMCMarketing - This a simple Customer Lifetime Value analysis using Buy Till You Die Modelling With PyMC Marketing library https://www.pymc-marketing.io