This project aims to enhance the targeting strategies for the Zalon campaign in "Space Pirates," a game developed by Creative Gaming. Through the application of uplift modelling, we identified customer segments most likely to respond positively to the campaign. This strategic approach allowed for more effective use of marketing resources, optimizing the adoption rate of the campaign.
The primary objective was to analyze data from 60,000 customers, divided equally into a control group and a treatment group, to identify the top 30,000 customers who would most benefit from targeted advertising. This was done to maximize campaign adoption and increase the efficiency of the marketing spend.
The project successfully targeted the optimal customer segment, resulting in a significant enhancement in the adoption rate of the Zalon campaign. By implementing uplift modelling techniques, we achieved a measurable increase in customer response rates and substantial ROI through targeted advertising efforts. Key performance metrics include Uplift (%) and Incremental Uplift (%).
cg-uplift.ipynb: Jupyter notebook containing the code for the uplift modelling analysis.data/: Directory containing the data sets used in the analysis.creative-gaming.pdf: Detailed case study PDF describing the project scope, methodology, and findings.
This project utilizes Python for data analysis, with libraries such as Pandas for data manipulation, Scikit-learn for applying machine learning techniques, and Matplotlib for visualizations.
- Clone the repository.
- Ensure you have Jupyter Notebook installed, or use Google Colab to open the .ipynb file.
- Install necessary Python packages: pyrsm, pandas, matplotlib, scikit-learn.
- Run the notebook cg-uplift.ipynb to see the analysis and results.