Abstract
Scoring leads manually can be time-consuming and ineffective because it relies on subjective criteria and human biases. On the other hand, automated lead scoring uses advanced algorithms to analyse large datasets and accurately predict a lead's potential value to the business. It provides valuable insights into customer behaviour and preferences that inform marketing and sales strategies. Machine learning algorithms are trained on historical sales data to identify patterns and correlations that indicate a lead's likelihood to convert. They consider variables like demographic information, web behaviour, and marketing engagement to generate a lead score that reflects the probability of converting into a customer. In this study, we conduct a comprehensive literature review on manual and automatic lead- scoring methods and the algorithms used in automatic lead-scoring. Additionally, we explore algorithms employed in automatic lead-scoring systems. Using PyCaret, a machine learning library, we compare various classification models to identify an effective lead- scoring model. Our results show that the GBC model achieved the highest accuracy of 82.82% with a precision of 78.55%.
Publication Date
5-2023
Document Type
Master's Project
Student Type
Graduate
Degree Name
Professional Studies (MS)
Department, Program, or Center
Graduate Programs & Research
Advisor
Ayman Ibrahim
Recommended Citation
Aziz, Ayad, "Lead Scoring using low code ML Library" (2023). Thesis. Rochester Institute of Technology. Accessed from
https://repository.rit.edu/theses/12217
Campus
RIT Dubai
