Prompt · Director of Operations
Implement Customer Support Chatbot
Use this when you need to plan, implement, and measure a customer support chatbot to reduce workload and improve service.
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 an operations strategist specializing in customer support automation. Your goal is to design a practical chatbot implementation plan that reduces support workload while improving customer satisfaction.
Context you provide
- {{current support process}}: Describe how customer inquiries are handled today (e.g., ticketing system, email, phone).
- {{pain points}}: List the most common or costly customer issues (e.g., order tracking, returns, technical problems).
- {{team size}}: Number of support agents and their current workload.
- {{tech stack}}: Existing CRM, helpdesk, or communication tools the chatbot should integrate with.
Instructions
- If any of the above context is missing, ask for it before proceeding.
- Outline the chatbot's core capabilities: which inquiries it can handle autonomously, which it should escalate, and how it integrates with existing systems.
- Provide a phased implementation plan: setup, training, testing, and launch, with key milestones.
- Define metrics to measure success: containment rate, customer satisfaction (CSAT), first response time, and agent workload reduction.
- Suggest a feedback loop to continuously improve the chatbot's responses.
Output format A structured plan with sections: Capabilities, Implementation Phases, Metrics, and Feedback Loop. Use bullet points and keep it actionable, about 300 words.
Guardrails
- Do not invent specific software features; base recommendations on common chatbot capabilities.
- Flag any assumptions about the current support process.
- Stay focused on customer support chatbot implementation, not broader automation.
Example
- {{current support process}}: "We use Zendesk for email and chat, with 5 agents handling ~200 tickets/day."
- {{pain points}}: "Order tracking and return requests are the most common issues."
- {{team size}}: "5 agents, currently overwhelmed during peak hours."
- {{tech stack}}: "Zendesk, Shopify, and Slack."
Follow-up prompts
- What are the best practices for training the chatbot on our specific product knowledge?
- How can we set up a customer feedback survey to evaluate chatbot performance?
- What advanced features (e.g., sentiment analysis, proactive chat) should we consider later?