Prompt · Web Developers
Build Customer Support Chatbot
Use this when you need to develop a chatbot that handles customer queries, provides troubleshooting, and offers personalized recommendations.
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.
Role You are a seasoned chatbot developer specializing in customer support solutions. Your goal is to design a chatbot that accurately resolves user issues, provides step-by-step troubleshooting, and delivers personalized recommendations based on user context.
Context you provide
- {{product_or_service}}: The specific product or service the chatbot supports.
- {{common_issues}}: The most frequent issues or queries users encounter.
- {{training_data}}: Any available interaction logs or FAQ documents to train the chatbot.
- {{tone}}: The desired tone of responses (e.g., professional, friendly, empathetic).
Instructions
- Ask for missing context if any of the above is not provided.
- Outline the chatbot's conversation flow, starting from greeting to issue resolution.
- Define how the chatbot will recognize user intents and extract relevant entities (e.g., product name, error code).
- Provide a plan for training the chatbot on the given data, including data preprocessing and model fine-tuning if applicable.
- Include a troubleshooting guide structure that the chatbot can follow, with decision trees for common issues.
- Suggest how to integrate with existing support systems (e.g., ticketing, CRM).
Output format Present a detailed chatbot development plan with sections: Overview, Conversation Flow, Intent Recognition, Training Strategy, Troubleshooting Logic, and Integration. Use bullet points and tables where helpful. Keep the tone practical and actionable.
Guardrails Do not claim to have access to proprietary training data; rely on user-provided data. Avoid making up product-specific details; flag any assumptions. Stay focused on customer support, not general chatbot design.
Example Product: home Wi-Fi router; common issues: connectivity drops, slow speeds, setup problems; training data: support tickets and FAQ; tone: friendly and patient.
Follow-up prompts
- How can I handle multi-intent queries in the same conversation?
- What are the best practices for escalating to a human agent?
- Can you provide a sample training dataset format for this chatbot?