X Education, an online education company, faces challenges with its lead conversion process despite generating numerous leads daily. With a typical lead conversion rate of only 30%, they aim to boost efficiency by identifying the most promising leads, termed 'Hot Leads', to focus their sales efforts effectively.Their lead conversion process resembles a funnel, where many leads enter but only a few become paying customers. Effective nurturing of potential leads in the middle stage is crucial for higher conversion rates.To address this issue, X Education seeks assistance in developing a lead scoring model. This model should assign a lead score to each lead, indicating their likelihood of converting into paying customers. The objective is to prioritize leads with higher scores for focused sales efforts, aiming to achieve a target lead conversion rate of 80%. In summary, X Education desires a model that can accurately predict lead conversion probabilities, facilitating targeted sales strategies and improving overall conversion rates.
1.Data Loading
2.Data Exploration (EDA)
3.Preprocessing
4.Feature Engineering
5.Outlier Analysis
6.Splitting Data into Train and Test Set
7.Model Building
8.Model Performance Benchmarking
9.Model Performance Evaluation
10.Cross Validation + Hyperparameter Tuning
11.Model Diagnosis Using Probability Calibration, ROC AUC Curve, Precision-Recall Curve
12.Making Predictions
13.Prediction on Test Set
14.Precision-Recall
By following these steps, X Education can develop a robust lead scoring system that helps prioritize leads with the highest conversion potential, ultimately improving the efficiency of their sales process.