A machine-learning application designed to predict whether a digital marketing lead is likely to convert into a customer.
How it works: Lead Data → Cleaning → EDA → Feature Analysis → ML Model → Prediction
What the project covers
- Data cleaning
- Exploratory Data Analysis
- Data visualization
- Feature analysis
- Machine learning classification
- Prediction
- Streamlit deployment
Business value
The concept demonstrates how businesses can use historical marketing data to identify patterns and support more data-driven lead prioritization.
The problem
Marketing teams generate many leads, but not all are equally likely to become customers, so it is hard to know where to focus.
The solution
A Streamlit app built with Python that takes lead data through cleaning, EDA, feature analysis, and a classification model to predict conversion.
What I learned
How much data cleaning and feature analysis shape what a model can actually learn.
What comes next
- Comparing additional models
- Showing which factors most influence conversion