AI has been achieved internally - The Journey
summary
Banca Intesa builds an internal AI-driven CRM: Python-based multiclass models produce next-best-product recommendations for 1.3 million customers and a separate pipeline for 120,000 SMEs, running omni-channel (mobile app, branch, call center, ATM, Viber/SMS/email via Infobip) with two‑way tracking. Models lift campaign conversion, contribute €300 million generated (≈65% of new loan production), and require continuous monitoring, data engineering (eightfold dataset growth), automated prospect enrichment, and churn-specific models to industrialize acquisition and maintain performance.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:08AI in modern banking
- 6:10AI in CRM and targeting
- 13:21Campaign optimization with ML
- 17:40Customer acquisition challenges
- 21:39Customer segmentation
- 23:23Churn prediction
- 27:41AI model development
- 30:11Measuring AI campaign performance
- 33:01Revenue impact of AI campaigns
- 35:12Strategic AI adoption
- 37:59Organizational transformations
- 40:50Early AI challenges
- 43:19AI winter and future trends
- 47:35Q&A
related talks
35mBank evolution to data science-driven business model
29mAI-Driven Business Success: Measurable Results and Revolutionized Sales
29mAI - Digital (r)evolution in banking
37mRole Of AI In Consumer Interaction
37mAI in the bancassurance paradigm
26mThe AI-Based Banking Era
37mApproaching AI in Banking
27mAI as a Service Scaling AI with Ease and Innovation