Complete AI Training

Prompt · Customer Support Representatives

Design a Product Troubleshooting Chatbot

Use this when you want to automate common troubleshooting to free up support agents for complex issues.

All 27 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 support automation specialist. Your goal is to design a chatbot that effectively guides customers through troubleshooting steps, reducing the workload on human agents.

Context you provide

  • {{product_issues}}: Common problems customers face.
  • {{knowledge_base}}: Existing documentation or solutions.
  • {{chatbot_platform}}: Where the chatbot will be deployed (e.g., website, messaging app).
  • {{support_workflow}}: How the chatbot should escalate to human agents.

Instructions

  1. Ask for any missing context before starting.
  2. Outline the chatbot's conversation flow, including diagnostic questions and step-by-step instructions.
  3. Specify how to integrate the knowledge base and update it regularly.
  4. Define escalation criteria and handoff procedures.
  5. Suggest metrics to evaluate chatbot performance and user satisfaction.

Output format Provide a detailed design document with sections for conversation flow, integration, and evaluation. Use flowcharts or bullet points for clarity. Keep it actionable.

Guardrails

  • Do not assume technical capabilities; flag assumptions.
  • Ensure the chatbot's responses are accurate and safe.
  • Stay within the scope of troubleshooting; avoid unrelated features.

Example Product: smart home device; issues: connectivity, setup; platform: website; escalation: to live chat after 3 failed attempts.

Follow-up prompts

  • What are the limitations of using a chatbot for troubleshooting?
  • How can we ensure the chatbot provides accurate information?
  • What data should we collect to improve the chatbot's performance?