https://github.com/rbhatia46/customer-lifetime-value-machinelearning
Computing Customer Lifetime value via Machine Learning approach
https://github.com/rbhatia46/customer-lifetime-value-machinelearning
Last synced: 7 months ago
JSON representation
Computing Customer Lifetime value via Machine Learning approach
- Host: GitHub
- URL: https://github.com/rbhatia46/customer-lifetime-value-machinelearning
- Owner: rbhatia46
- Created: 2021-11-13T07:56:34.000Z (almost 4 years ago)
- Default Branch: main
- Last Pushed: 2022-02-03T15:21:16.000Z (over 3 years ago)
- Last Synced: 2025-01-24T18:37:07.818Z (9 months ago)
- Language: Jupyter Notebook
- Size: 643 KB
- Stars: 1
- Watchers: 3
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Customer-Lifetime-Value-MachineLearning
## 3 Questions to answer -
1. Which customers have the highest spend probability in next 90 days ?
2. Which customers have recently purchased but unlikely to buy ? (look at customers have bought anything in last 90 days but have a lower purchase probability as per the model(less than 20%)) - revive the customer before they die
3. Which customers were predicted to purchase but didn't(missed opportunities) ? (people who were predicted to spend a certain amount and had a higher purchase probability in the next 90 days but actually spent 0 dollars) - get the marketing team to send these people targeted emails because these are missed opportunities that can boost the revenue quite significantly
## Next Steps for improvement -
1. Leverage More features rather than just RFM (Customer demographics, Geographic information, etc)
2. Try better models(maybe AutoML) and do a more in depth hyperparameter tuning
3. Apply Interpretable ML techniques like SHAP and LIME to dive deeper and explain the predictions
4. Predict the next purchase day
5. Integrate through a REST API to create a dashboard.
6. Leverage Product/Item catalog(if available)
7. Try a Pareto/NBD Model and other classic statistical approaches.