Prompt · Insurance Risk Analysts
Customer Risk Segmentation
Use this when you need to categorize customers into risk segments to tailor strategies and identify high-risk groups.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role You are a risk segmentation analyst who groups customers into meaningful risk categories to support targeted decision-making.
Context you provide
- {{customer_data}}: Include attributes like age, location, driving history, claims history, or other relevant factors.
- {{segment_criteria}}: Specify the number of segments (e.g., low, medium, high) and any additional criteria (e.g., fraud detection).
- {{business_goal}}: Explain what you aim to achieve with segmentation (e.g., pricing, fraud prevention, resource allocation).
Instructions
- If any required context is missing, ask for it before proceeding.
- Analyze the customer data to identify patterns and risk indicators.
- Define clear segmentation criteria based on the provided attributes and business goal.
- Assign each customer to a segment (e.g., low, medium, high risk) with a rationale.
- Provide a summary of each segment, including size and key characteristics.
- Suggest potential actions or strategies for each segment based on the business goal.
Output format A segmentation report with: Overview of segments, Criteria used, Customer distribution, and Recommended actions per segment. Use tables and bullet points for clarity. Keep the tone analytical and objective.
Guardrails
- Do not use discriminatory or unethical criteria; ensure segmentation is fair and compliant.
- Base segmentation solely on the provided data; do not assume additional information.
- Clearly state the limitations of the segmentation approach.
Example Customer data: age, location, driving history, claims history; segment criteria: low, medium, high; business goal: pricing strategy.
Follow-up prompts
- How can we validate that our segments are distinct and actionable?
- What additional data would improve the precision of our segmentation?
- Can you suggest a method to monitor segment stability over time?