This project builds a machine learning model that predicts the conversion probability of leads for an educational technology company. By assigning a predictive lead score, the sales team can effectively prioritize their efforts on high-potential prospects, thereby increasing efficiency and boosting conversion rates.
The analysis was performed on the Lead Scoring dataset from Kaggle. Link to Dataset
- Exploratory Data Analysis (EDA): Investigated the relationships between lead attributes and the final conversion status.
- Data Preprocessing: Cleaned the data by handling missing values and encoded categorical variables using one-hot encoding.
- Model Building: Developed and compared three classification models: Logistic Regression, Random Forest, and XGBoost.
- Model Evaluation: Assessed models based on their AUC-ROC score, Accuracy, and Precision to select the best-performing algorithm.
-
Finding 1: Leads who spend more time on the website have a significantly higher conversion rate. This is the single most important behavioral metric for predicting conversion.
-
Finding 2: The lead's origin is a powerful predictor. Specifically, leads generated through the "Lead Add Form" convert at a very high rate and should be prioritized.
The XGBoost classifier demonstrated the best performance across all key metrics.
- Clone this repository.
- Install the required libraries:
pip install pandas numpy matplotlib seaborn scikit-learn xgboost jupyter - Open and run the
lead_scoring_analysis.ipynbJupyter Notebook.