This project compares Facebook Ads and Google Adwords to determine which platform performs better in terms of conversions, clicks, and cost efficiency. The project uses A/B testing, statistical analysis, regression, and time-based trends to generate insights.
- Identify which ad platform performs better.
- Analyze the relationship between clicks and conversions.
- Measure the frequency of high-conversion days.
- Conduct A/B hypothesis testing.
- Study day-based and time-based trends.
- Evaluate how ad spend impacts conversions.
- Determine which campaign provides better ROI.
H0: Both platforms have equal average conversions. H1: Facebook has higher average conversions than Adwords. Statistical results show Facebook performs better, so the null hypothesis is rejected.
Conversions and clicks show near-normal distribution with few outliers.
Facebook has more high-conversion days and stronger engagement.
Clicks and conversions have a strong positive correlation (0.87).
Ad spend, especially on Facebook, strongly predicts conversions.
Conversions remain consistent across weekdays, with a small rise mid-month.
Statistical results confirm conversions depend significantly on ad spend.
- Loaded and reviewed the dataset.
- Cleaned and prepared the data.
- Performed descriptive statistics.
- Conducted exploratory data analysis.
- Compared platform performance.
- Performed A/B and hypothesis testing.
- Built a regression model.
- Checked variable correlations.
- Analyzed time-based patterns.
- Summarized insights and conclusions.
Python (Pandas, NumPy, Matplotlib, Seaborn), SciPy, Statsmodels, Jupyter Notebook
project/ data/ ad_data.csv
notebook/ AB_Testing_analysis.ipynb
README.md
requirements.txt