Watch Step By Step Video Tutorial Below
1. Create a WhatsApp Chatbot Using Can and Not
To create a WhatsApp chatbot, you need to utilize Can and Not integrations effectively. Start by opening your WhatsApp and sending a message to your chatbot. For example, you might ask, ‘Which apartment or villa do you have available within my budget?’
Once you send your message, the AI agent will respond with a confirmation. It may say, ‘Thanks, just to confirm, you mean a budget of this much.’ This interaction is a direct result of the Can and Not integrations working together to process your inquiry.
2. Setting Up Pabi Chatflow for Integration
To set up your chatbot, navigate to Pabi Chatflow. You need to sign in or sign up for a new account. If you are a new user, click on the ‘Sign Up for Free’ option to receive credits for testing. Existing users can simply sign in using their credentials.
- Click on ‘Add WhatsApp Number’ to connect your WhatsApp.
- Choose the preferred method to connect, either through WhatsApp Connect or a manual token.
- Follow the instructions to complete the connection.
After connecting your WhatsApp, you will see the dashboard displaying your credits and connected WhatsApp number. This setup is crucial for ensuring your chatbot can communicate effectively with users.
3. Creating Your AI Assistant with Can and Not
To create your AI assistant, click on the ‘AI Assistant’ option in the Pabi Chatflow dashboard. Name your assistant, for example, ‘WhatsApp AI Agent.’ This naming helps in identifying the assistant later.
Once named, you will configure your assistant. Set the instruction type as AI agent and adjust the creativity level. For instance, if you set it at 0.7, the AI will provide more creative responses. You will also need to enter the API key from OpenAI to enhance the AI’s capabilities.
- Click on ‘Create New Secret Key’ on the OpenAI API page.
- Provide a name for the key and create it.
- Copy the API key and paste it in your assistant settings.
This configuration allows your assistant to respond to inquiries effectively, making it more interactive and useful for users.
4. Setting Knowledge Source and Assistant Settings
For your AI assistant to provide accurate responses, you need to upload a detailed knowledge source. This can be a document containing all relevant information about your services. Select ‘Upload File’ in the knowledge source section and choose your document.
Additionally, configure your assistant’s settings by entering header and footer messages. You can toggle on or off the display of these messages based on your preferences. Make sure to include a fallback message for when the AI cannot provide an answer.
- Add a fallback message such as, ‘I am here to assist with questions related to company information.’
- Set a stop keyword for users who want to escalate their queries to human support.
- Adjust retry attempts for the AI to ensure it tries multiple times to respond.
These settings will greatly enhance the user experience, ensuring that your AI assistant is both informative and responsive.
5. Assigning Your AI Assistant to Contacts
Once your AI assistant is configured, you can assign it to specific contacts or groups. Navigate to the inbox settings and select the AI auto-reply settings to enable your assistant.
You can create multiple rules to assign the assistant to various groups or all contacts. Ensure you save these rules after setting them up. This step is essential to ensure your assistant is actively responding to inquiries.
- Select the AI assistant you created from the list of assistants.
- Add rules for different contact groups as needed.
- Click ‘Save’ to confirm your settings.
By assigning your AI assistant correctly, you ensure that it can handle inquiries efficiently, improving customer satisfaction and engagement.
Conclusion
In this tutorial, we explored how to create a WhatsApp chatbot using Can and Not integration. By following the steps outlined, you can automate your conversations effectively, enhancing user interaction and satisfaction.
