Key results
improvement in cross-sell conversions compared to random targeting using machine learning-based segmentation
accuracy in detecting high-propensity customers for cross-sell and up-sell offers
uplift in monthly premium income by engaging the most relevant customers with tailored offers
reduction in manual reporting efforts through automated dashboards for sales, onboarding, and servicing
About the Client
A top-tier Indian life insurance company offering term, ULIP, health, and savings plans to millions of policyholders, with a robust multi-channel distribution network spanning agency, bancassurance, and digital platforms.
Key Challenges
Ineffective Audience Targeting - Broad, demographic-based campaigns resulted in low response rates and irrelevant product messaging.
Limited Customer Intelligence - Fragmented data across policy, CRM, and service systems made it difficult to identify up-sell or cross-sell opportunities.
No Lead Prioritization or Propensity Scoring - Sales and call center teams lacked tools to focus on high-value or high-intent leads.
Manual, Delayed Sales Reporting - Reliance on Excel-based reporting caused delays in decision-making and lack of real-time visibility.
Our Approach
Unified Data Foundation
- Integrated policy, claims, CRM, and onboarding data into a centralized Azure Synapse warehouse
- Built unified customer profiles using entity resolution techniques
- Created historical and near real-time models to power analytics and ML
Sales & Marketing Intelligence
- Developed Power BI dashboards covering sales performance, agen productivity, product-wise penetration, and onboarding drop-offs
- Enabled territory-wise drilldowns and trend views for management
- Automated performance alerts and KPI tracking for actionable insights
Propensity Modeling for Cross-Sell & Up-Sell
- Built machine learning models in Python to predict customer affinity to additional products
- Used transactional, behavioral, and demographic features for scoring
- Integrated scores into CRM for lead prioritization and targeting
Personalized Campaign Activation
- Segmented customer base by product affinity, CLTV, and risk appetite
- Integrated campaign triggers via email, SMS, and agent workflows
- Created ROI dashboards to monitor campaign success, Cost per lead (CPL), and conversions
Implemented Solution
Data Warehousing Layer
- Built on Azure Synapse with scalable architecture
- Automated ingestion via Azure Data Factory and Python pipelines
- Designed star schema models for policy, customer, claims, and transactions
BI & Reporting Dashboards
- Delivered 15+ dashboards for sales, servicing, agent performance, onboarding, claims, and retention
- Role-based access for CXOs, regional sales heads, and agents
- Real-time performance monitoring and drill-down features
ML & Predictive Analytics Layer
- Developed and deployed models using Python
- Automated weekly scoring pipeline with continuous retraining
- Propensity scores pushed into CRM and campaign management platforms
Campaign Execution
- Triggered outreach based on model-driven lists using SMS, email, and call center CRM
- Connected with tools like Netcore or CleverTap for automated campaign delivery
- Tracked campaign impact via Power BI dashboards by channel, product, and region
Technologies Used
