This project schedules a smart and simple promotional strategy for Starbucks where firstly we analyse simulated data of the Starbucks Rewards mobile app and then we prepare machine learning models to predict individual offer portfolios to each customer.
It asks and answers following business questions:
- Can we find relationships between customer demographic group age and income?
- How are offer types distributed in data set related to customer demographics age, income and gender?
- Which are the most popular offers?
- Are there significant differences in age, income and registration date for promotion success?
- Can we build a classification model to predict promotional offer success for new customers based on their demographical data age, income, registration date and gender?
A blog post with the results can be found on medium.
Jupyter notebook environment based on python 3.8 with libraries pandas, numpy, json, datetime, joblib, matplotlib.pyplot and sklearn
Download zip folder on local computer, extract files and open "Starbucks_Capstone_notebook.ipynb" with jupyter notebook.
Below you can find the file content of this project:
- data
|- portfolio.json (json file with data to process)
|- profile.json (json file with data to process)
|- transcript.json (json file with data to process)
- LICENSE (file with MIT licence)
- README.md (markdown file with instructions)
- Starbucks_Capstone_notebook.ipynb (jupyter notebook file with analysis)
- Starbucks_Capstone_notebook.html (html file with analysis)
This project is licensed under the terms of the MIT license
Author: Eugen Iftimoaie
For questions feel free to contact me on my e-mail adress: eugen.iftimoaie@gmx.de