Supporting the Edtech Industry: Optimizing Marketing Budget Through Data Driven Client Insights

Objective

The company is an initial stage startup that offers programs on cutting-edge technologies to students and professionals to help them upskill/reskill. With a large number of leads being generated on a regular basis, one of the issues faced by the startup is to identify which of the leads are more likely to convert so that they can allocate resources accordingly. We have been provided the leads data to:

• Analyze and build an ML model to help identify which leads are more likely to convert to paid customers,

• Find the factors driving the lead conversion process

• Create a profile of the leads which are likely to convert

Approach:

Using EDA Techniques we will see the univariate and multivariate relationships in the data and gather insights after which we will use a Decision Tree classifer model based on the objective. We will likewise examine the feature importance to identify what features most strongly indicate the likelihood of conversion; and these steps will give a comprehensive guide to the startup on where best to allocate its marketting budget to obtain the highest return on investment.

Summary of Findings:

Time spent on the website emerged as the most significant driver, with higher conversion rates linked to users who engage deeply with the website, complete their profiles (75%-100%), and spend over seven minutes on the platform. Three target personas were identified: Group A (24+ years, website visitors preferring phone communication), Group B (professionals with high-profile completion and website engagement), and Group C (students with high website activity). Emails proved the most effective engagement channel, while chatbots and online communication tools also contributed to conversions. Although the mobile app and referrals showed limited impact on conversions, digital media and educational channels remain critical for creating awareness and driving traffic.

Key Data Science concepts covered:

Decision Trees

Random Forest

Hyperparameter Tuning

Exploratory Data Analysis

Click the link below to download the full code report.

Potential+Customers+Prediction (1).pdf

Download PDF • 2.99MB