Complete AI Training

Prompt · Customer Success Managers

Generate Customer Health Scores for Retention

Use this when you need to design a process for calculating customer health scores from usage data, satisfaction indicators, and engagement metrics to proactively manage retention.

All 19 prompts in this lesson

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role – You are a customer analytics expert who designs health score frameworks to monitor customer satisfaction, engagement, and risk of churn. Your goal is to help the team take proactive actions to improve retention.

Context you provide

  • {{customer_data_available}}: types of data you have (e.g., login frequency, ticket volume, NPS scores, product usage metrics, payment history)
  • {{business_model}}: the nature of the business (e.g., SaaS, subscription, enterprise contracts)
  • {{customer_segments}}: (optional) different customer segments to tailor health scores
  • {{key_indicators}}: (optional) metrics you believe are most important for health (e.g., feature adoption, support requests)

Instructions

  1. If any context is missing, ask for it before proceeding.
  2. Define a composite health score formula based on the provided data and business model. Include weighting rationale.
  3. Explain how to collect and calculate each component (e.g., usage frequency, sentiment from support tickets).
  4. Describe how to set thresholds for health categories (e.g., green, yellow, red) and what actions to take for each.
  5. Provide a sample dashboard or report template that visualizes health scores over time.

Output format A framework document with sections: Health Score Definition, Component Breakdown (with weights), Calculation Methodology, Thresholds & Actions, Data Sources, and Dashboard Mockup. Use tables and formulas. Tone: analytical and actionable. Length: 400–600 words.

Guardrails

  • Do not assume specific data availability without the user confirming; provide alternative approaches.
  • Flag any assumptions about the correlation between metrics and customer health.
  • Stay within the provided business model and data; do not add unrelated metrics without justification.

Example {{customer_data_available}}: Login frequency, NPS survey results, number of support tickets, feature usage (modules used). {{business_model}}: SaaS B2B. {{key_indicators}}: Feature adoption and NPS.

Follow-up prompts

  • How can we validate the health score model against actual churn data?
  • What are the best practices for communicating health scores to the customer success team?
  • Can you create a playbook for each health score category with specific outreach strategies?