Leveraging Retail Retention Strategies Using AI-Driven Predictive Customer Insights

Kudupudi, Lakshmi Priya (2025) Leveraging Retail Retention Strategies Using AI-Driven Predictive Customer Insights. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Customer churn remains a major concern for retailers, where even minor service issues can trigger customer loss. This study develops a scalable, AI-driven churn prediction and prevention system integrating cloud storage, predictive modeling, CRM automation, and business dashboards. Using the 90,000+ record Olist e-commerce dataset, Data was stored in Azure Blob Storage, processed in Jupyter Notebook, and modeled using Naive Bayes, KNN, XGBoost, and Bi-LSTM. XGBoost and Bi-LSTM achieved perfect accuracy and ROC-AUC, with Bi-LSTM excelling at sequence-based behavior analysis. High-risk customers i.e, churn score above 0.80 were passed on to Salesforce via REST API, with custom objects, Apex Triggers enabling automated follow-up and email notifications. Power BI dashboards aggregated Azure and Salesforce data for real-time churn monitoring. Results show that integrating AI predictions with Salesforce significantly enhances proactive retention, yielding a practical, reproducible model for operationalizing churn prevention in real CRM environments.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Agarwal, Bharat UNSPECIFIED
Uncontrolled Keywords:	Customer Churn; Predictive Analytics; Bi-LSTM; XGBoost; Salesforce CRM; Azure Blob Storage; Power BI; Machine Learning; Deep Learning; Retail Analytics
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence H Social Sciences > HF Commerce > Marketing > Consumer Behaviour H Social Sciences > HD Industries. Land use. Labor > Specific Industries > Retail Industry
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	25 Aug 2026 15:40
Last Modified:	25 Aug 2026 15:40
URI:	https://norma.ncirl.ie/id/eprint/9638

Actions (login required)

View Item