The problem
Customer reviews arrive as free text and pile up in a queue. Someone has to read each one to decide what it is about, how upset the customer is, who should handle it, and whether this customer is about to leave. The last question is the one that costs money, and it is the one a keyword filter cannot answer.
How it works
The review's text and star rating go into the where clause of four _predict calls that run in parallel: category, sentiment, assignee, and churn within 90 days. Each returns its top value with confidence, alternatives, and $why, so the reviewer sees which words carried the prediction.
The churn prediction is what connects feedback to retention: the same review that is routed to the right person also flags the customer for the retention queue.
For the full architecture, see the technology overview. For the broader narrative across multiple use cases, read The Predictive Application.
See it live
This use case runs in the 🛒 E-commerce demo today. Click through to the live application and inspect the queries that produce the result.
