The graphs which are plotted using Plotly cannot be displayed in Github. Visit https://nbviewer.jupyter.org/github/tejaslinge/Marketing-Analysis/blob/main/Analysis.ipynb to view the complete Jupyter notebook.
This project demonstrates how to perform EDA on marketing campaign data and make future marketing performance predictions.
Calculations of marketing metrics CTR, CPC, CPA, Conversion rate, Subscription rate, ROI for campaign as a whole and for each individual adgroups is shown in this notebook. Along with performance metrics, performance of the campaign for each adgroup should also be explored to find optimal marketing solutions i.e the groups which the next marketing campaign should focus on to increase the ROI and conversion rate.
To make future predictions of performance of the campaign, I've implemented Decision Trees regressor, with error functions RMSE and R-squared value to evaluate the performance of our model. We can further perform more EDA on the dataset, and find which ads deliver the highest ROI in each adgroups.