This project focuses on analyzing the effectiveness of two marketing campaigns — a control group and a test group — to determine which one delivered better performance.
I was tasked with evaluating the success of these campaigns using a combination of exploratory data analysis (EDA) and statistical testing. The main objective was to assess whether the test campaign produced a statistically significant improvement over the control.
- Exploratory Data Analysis (EDA): Initial visual and statistical summaries were used to understand the structure and behavior of both groups.
- Two-sample t-test: A statistical t-test was performed to compare the conversion rates between the control and test groups and validate whether the difference was significant.
- KPI Calculation – ROI:
- Since actual revenue per conversion wasn't available, a hypothetical value was assumed to estimate revenue and calculate Return on Investment (ROI).
- ROI was derived using reach, conversion rate, spend, and assumed conversion value.
- This provided a more holistic view of each campaign's efficiency beyond conversion rate alone.
The analysis demonstrated that:
- The test group had a statistically higher conversion rate than the control group.
- However, when evaluating ROI, the test group did not necessarily outperform the control group, due to differences in reach and spend.
- These combined insights provided a more nuanced picture of campaign performance and helped identify the more effective marketing strategy.
- Python (Pandas, NumPy, SciPy, Seaborn, Matplotlib)
- VS Code
Feel free to reach out with questions or feedback!