At-Risk Customer Intervention Plan
The prompt
You are a customer success manager building an intervention plan for a Red-status account. Account data: [PASTE: Account | ARR | Health indicators (usage drop / support escalations / NPS low / champion left / payment issue) | Renewal date | What has been tried | Root cause hypothesis] Build the intervention plan: 1. Root cause — what is actually driving the risk? (product gap / adoption failure / competitive / relationship / budget) 2. Intervention owner — CSM / account manager / VP-level / executive sponsor? 3. Specific actions — each action tied to a root cause; not generic "check in calls" 4. Timeline — what must happen in the next 7 / 30 / 60 days to prevent churn? 5. Go/no-go decision — at what point do we accept churn is likely and shift to minimum-cost retention vs. maximum-effort recovery? Output: Account intervention plan. Day 1 / Week 1 / Month 1 actions with owner. Decision trigger for escalation or accept.
Why this works
Requiring a root cause hypothesis upfront prevents the intervention plan from being a list of actions without a diagnosis. The four intervention phases (acknowledge / stabilise / recover / prevent) match the psychological sequence of a customer rescue — jumping straight to 'here's our plan' before acknowledgment often makes things worse. Escalation criteria built into the plan prevents intervention delays when the situation deteriorates.
Risks & review
Root cause diagnosis is where most at-risk plans fail — it's often easier to attribute churn to 'product gaps' than to the CS team's adoption work. Be honest in the root cause section, including any internal factors. The AI will generate a logical intervention plan; the human judgment required is whether the intervention is actually addressing the real reason the customer is at risk.