South KoreaRetail and e-commerce2024 – ongoingJeongyookgak
Churn prediction and automated CRM campaigns for an online meat retailer
Jeongyookgak raised its active customers by 30% with churn prediction and automated CRM messages
Problem
Jeongyookgak, a Korean online and store-based meat retailer, kept its data in a legacy Oracle cloud data mart with silos and little flexibility. Staff waited about three days for data requests. Marketing had no reliable way to spot customers about to stop buying.
Approach
The company consolidated data from several cloud sources onto Databricks and built a churn prediction model that analyzes customers' purchase cycles. The system sends automated CRM messages to customers during periods of churn risk. It also used the data for demand prediction and inventory management.
Results
Takeaway
Timing messages to each customer's purchase cycle gave the retailer more return from push marketing. The page gives no baseline for the active-customer figure.
Categories
- Industry
- Retail and e-commerce
- Business function
- Marketing and sales
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks describes how Jeongyookgak, which operates the Chorocmaeul brand in Korea, moved from a legacy data mart to Databricks and built a churn prediction model with automated CRM messages. It reports a 30% increase in active customers, 2x returns on push marketing, a 10% retention improvement and a tripled conversion rate after coupon-reminder campaigns.
Full details are in the original sources.
