churn-analystlisted

# Churn Analyst Generic retention tactics (discount offers, "we miss you" emails) underperform because they treat all churn as the same problem. Real churn analysis segments by *why* customers left, since the fix for "never onboarded successfully" is completely different from "found a cheaper competitor" or "outgrew the product." ## Workflow 1. **Get the data**: churn events with timestamps, and ideally account attributes (plan tier, tenure, usage/engagement metrics, support ticket history) alongside them. Read attached data directly rather than asking the user to describe it. 2. **Segment churn by tenure first.** Early churn (within the first 30-90 days, roughly) almost always indicates an onboarding/activation problem — the customer never reached the point of experiencing real value. Late churn (established, engaged accounts leaving) indicates a different problem — competitive pressure, changing needs, or a specific negative experience. Don't propose the same fix for both. 3. **Look for leading indicators before the churn event itself** — declining usage/login frequency, a drop in a key engagement metric, a support ticket that went unresolved, a missed renewal conversation. These are what make churn *predictable* rather than only explainable after the fact, and they're what a retention intervention should actually trigger on. 4. **Correlate churn rate against account attributes** — plan tier, acquisition channel, company size/segment — to find which sub-populations churn