Intuit-Quickbooks-Upgrade-Targeting-Model

Intuit QuickBooks Upgrade Analysis

Project Overview

This project aimed to develop a predictive model to identify small businesses likely to respond to an offer to upgrade to QuickBooks version 3.0. Utilizing data from a wave-1 mailing campaign, our goal was to maximize the effectiveness of Intuit's wave-2 mailing campaign. Advanced predictive modeling techniques, including Logistic Regression and Neural Networks, were employed to analyze business interactions, purchasing history, and prior campaign responses.

Objective

The primary objective was to refine the selection process for the wave-2 mailing, aiming to increase response rates and maximize the campaign’s overall effectiveness and profitability. This involved evaluating different classifiers, performing extensive hyperparameter tuning, and enhancing model accuracy.

Repository Contents

1. intuit.ipynb: Jupyter notebook containing the main analysis and modeling code.
2. data/: Data sets used in the analysis.
3. intuit-quickbooks-case.pdf: PDF document detailing the case study.
4. intuit_analysis.ipynb: Additional explanations and write-up of the analysis.
5. to_target_businesses.csv : Data file containing the list of businesses to target in the wave-2 campaign.

Technologies Used

1. Predictive Modeling: Logistic Regression, Neural Networks
2. Performance Metrics: AUC, Confusion Matrix, Gains and Lift Charts
3. Model Tuning: Hyperparameter Optimization, Cross-Validation
4. Economic Analysis: Breakeven Analysis, Campaign Targeting Strategy

How to Run

1. Clone the repository.
2. Ensure you have Jupyter Notebook installed to open the .ipynb files.
3. Install necessary Python packages: pandas, numpy, scikit-learn, matplotlib, pyrsm.
4. Run the notebook intuit_code.ipynb to view the analysis and modeling steps.