This project provides a comprehensive analysis of marketing campaign efficiency using Python and the Pandas library. The goal is to evaluate the acquisition costs versus the revenue generated across various dimensions like media channels, platforms, and geographic locations.
The analysis is based on two primary datasets:
in_app_events: Contains application installation and payment events for users acquired through specific marketing campaigns.cost: Contains the marketing expenditures (spend) associated with those campaigns.
The analysis was performed in three stages:
Calculated the core results for each campaign:
- Unique installs and unique payers.
- Total marketing costs.
- Net and Gross revenue.
- Total number of successful payment events.
To measure success, the following five metrics were calculated:
- CPI (Cost Per Install)
- CR% (Conversion Rate from Install to Payer)
- ARPU (Average Revenue Per User/Install)
- ARPPU (Average Revenue Per Paying User)
- ROAS% (Return on Ad Spend)
Based on the data processing, the following segments demonstrated the highest efficiency:
- Top Media Channel: Facebook Ads is the leader in both total revenue and ROAS (7.4%). However, further cost optimization is recommended to improve margins.
- Top Platform: Android outperformed other platforms, securing the highest revenue and a ROAS of 9%.
- Geographic Leaders:
- DK (Denmark) achieved the highest overall ROAS (14%) and ranks second in total revenue.
- US (United States) is the absolute leader in revenue and maintains a strong second place in ROAS (9.3%).