Business Situation
- A global media and entertainment company managing digital, TV, and streaming platforms faced significant challenges with high advertiser churn rates.
- The lack of predictive insights into advertiser behavior and fragmented engagement metrics across multiple platforms led to inefficient retention efforts and difficulties in revenue forecasting.
SGA Approach
Technology
- Implemented a centralized data lake based on Google Cloud Platform (GCP) for holistic insights.
- Integrated churn propensity scores with Salesforce CRM.
AI
- Developed machine learning models for churn prediction.
- Implemented AI-driven churn propensity scoring.
- Utilized gradient-boosting decision trees, long short-term memory (LSTM) models, and survival analysis techniques.
Research
- Designed a comprehensive churn framework covering slow, sudden, early, and flatline churn behaviors.
Key Takeaways
- Reduced advertiser churn by 5%, retaining key clients.
- Developed a comprehensive churn prediction framework covering various churn behaviors.
- Integrated churn propensity scores directly into Salesforce CRM for proactive engagement.
- Enhanced revenue forecasting capabilities with more accurate churn predictions.
- Streamlined sales processes by prioritizing high-risk accounts within Salesforce workflows.
