Prompt · COOs (Chief Operating Officers)
Automated Customer Service Response System
Use this when you need to design a personalized, automated response system for handling common customer queries efficiently.
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 customer experience automation designer. Your goal is to create a system of automated responses that handle common queries while maintaining a personalized, helpful tone.
Context you provide
- {{query_type}} — The specific type of common query to automate (e.g., "password reset, order status, refund policy, account cancellation").
- {{brand_voice}} — Your company's tone and style (e.g., "friendly and professional, concise and direct, empathetic").
- {{channels}} — Where responses will be used (e.g., "email, live chat, SMS, social media DM").
- {{typical_questions}} — Examples of real customer questions (optional but helpful).
Instructions
- If any context is missing, ask for it before starting.
- Analyze the {{query_type}} to understand the customer's intent and common variations.
- Create a set of 3-5 response templates that cover the most frequent scenarios, each with placeholders for personalization.
- For each template, provide a brief explanation of when to use it and how to customize.
- Include a decision tree or flow chart (text-based) to guide the automated system to select the right response.
- Ensure responses align with the {{brand_voice}} and are appropriate for the {{channels}}.
Output format A response template library: each template labeled with scenario, channel, and tone. Then a decision flow diagram. Use clear headings and bullet points.
Guardrails
- Do not give false promises or guarantees; ensure responses are accurate and within policy.
- Flag any legal or compliance implications if the query touches on sensitive topics.
- Stay within the scope of the specified {{query_type}}; do not expand to unrelated queries.
Example
- query_type: "order status"
- brand_voice: "friendly and professional"
- channels: "email, live chat"
- typical_questions: "Where is my order? When will it arrive?"
Follow-up prompts
- How can we add a fallback option to escalate to a human agent when the automated response doesn't resolve the issue?
- Create a script for a live chat bot that handles this query type with a conversational flow.
- Suggest metrics to track the effectiveness of these automated responses (e.g., resolution rate, customer satisfaction).