Agentic AI for Post2Pre Churn Prevention
Business Problem: Postpaid customers downgrading to prepaid plans causes immediate ARPU loss and reduced customer lifetime value. Traditional rule-based retention systems cannot process unstructured data, operate in real-time, or personalize interventions.
Agentic Solution: 7 specialized AI agents autonomously analyze 5 datasets to detect churn risk early, diagnose root causes, design personalized Floor/Middle/Ceiling offers, conduct empathetic conversations, and measure intervention effectiveness. Targets customers with AON âĨ 90 days.
7
AI Agents
5
Datasets
18
Agent-Data Links
90+
Days AON Threshold
Real-time
Processing Mode
100%
Automation Rate
| Capability | Traditional Rules | Agentic AI |
|---|---|---|
| Churn Detection | â Batch processing (daily/weekly) | â Real-time risk scoring |
| Data Analysis | â Structured data only (ARPU, tenure) | â Multi-modal (profiles + transcripts + sentiment) |
| Root Cause | â Not identified | â NLP extracts reasons from conversations |
| Interventions | â Generic discount for all | â Personalized Floor/Middle/Ceiling offers |
| Conversations | â Limited by human agent capacity | â Scalable LLM-powered dialogue (Arabic + English) |
| Adaptation | â Manual rule updates | â Continuous learning from outcomes |