ArgentinaFinance and insurance2019 – ongoingBanco Macro
IBM Watson Studio machine learning for churn, next best action and marketing
Banco Macro's commercial intelligence team built churn and next-best-action models on IBM Watson Studio and PowerAI
Problem
The bank wanted to anticipate what products customers would take and which customers might leave. Its data was sensitive and needed a controlled environment.
Approach
A four-person business intelligence team ran IBM Watson Studio on premises on IBM PowerAI infrastructure, using open-source data science tools. The models predicted product uptake, the next best action for existing clients and churn over a five-year horizon. IBM Watson Campaign Automation delivered personalized campaigns. Fraud detection was being added step by step.
Results
No published figures
Takeaway
The bank chose an on-premises set-up to keep customer data in a controlled environment. The interview reported no results.
Categories
- Industry
- Finance and insurance
- Business function
- Marketing and sales
- Innovation type
- Process innovation
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Company's own claim
Sources
Source summaries
Interview-based article with the bank's commercial intelligence team. It describes the IBM Watson Studio and PowerAI set-up, the use cases (product propensity, next best action, churn, fraud) and the hardware. It says the team began building predictive models about three years earlier and reports no results.
Full details are in the original sources.
