Objective: Integrate Bevatel with Dialogflow to leverage chatbots for automated responses and seamless agent handoffs, enhancing customer support capabilities.

Why Integrate Bevatel with Dialogflow?

Chatbots are an essential component of any customer service platform. When handling large volumes of conversations, scaling human support may not be the most efficient solution. By integrating Dialogflow with Bevatel, you can deploy a chatbot that answers common questions and hands over complex queries to an agent, optimizing your customer support workflow.

Key Benefits of Dialogflow Integration:

• Automate responses to FAQs.
• Improve response times with chatbot handling.
• Seamless handoff to live agents when required.
• Support for interactive messages and advanced intents.
  
  

Configuring Dialogflow Integration in Bevatel

Prerequisite: Ensure you have a Dialogflow bot set up before configuring the integration in Bevatel. If not, follow the steps in the next section to create a Dialogflow bot.

Steps:

1. Go to Settings → Integrations → Dialogflow in the Bevatel dashboard.
2. Click on Configure to start the setup.
3. Add the following:

• Project ID: The unique identifier for your Dialogflow project.
• Credentials: Paste the JSON key contents you downloaded from Google Cloud (details on how to create these credentials are explained in the "Create a Service Account" section below).
• Select Channel: Choose the appropriate inbox or channel for chatbot interactions.

4. Click Create to finalize the integration.

Creating a Dialogflow Bot for Bevatel

If you don't have a Dialogflow bot yet, follow these steps to create one:

1. Create a New Agent

1. Go to the Dialogflow Console.
2. Click on Create new agent and fill in the required details such as agent name and language.
3. Click Create.
   
   
   
   
   

2. Create Intents

1. In your Dialogflow agent, create intents to define how your bot should respond to specific queries.
2. You’ll see two default intents: "Default Fallback Intent" and "Default Welcome Intent."
3. You can add more intents to suit your use case. For example, you can create an intent to greet customers or answer specific FAQs.

3. Test the Bot

Test your bot by interacting with it directly on Dialogflow. Ensure the responses align with the intents you created.

Connecting Dialogflow Bot to Bevatel

To connect the Dialogflow bot with Bevatel, you need to create a service account in your Google Cloud Console.

1. Create a Service Account

1. Go to the Google Cloud Console.
   
   

2. Navigate to IAM & Admin → Service Accounts.
   
   

3. Click Create Service Account, provide a name and description, and click Continue.
   
   

4. Under Roles, choose Dialogflow API Client.
   
   

5. Click Done.

2. Create and Download Key

1. Once the service account is created, select it from the list.
2. Navigate to the Keys tab and click Add Key.
3. Choose JSON and download the key file. This key file is necessary for connecting Dialogflow to Bevatel.

3. Add Key to Bevatel

1. Go back to the Bevatel integration settings.
2. In the Credentials field, paste the contents of the JSON key file you downloaded earlier.
3. Click Create to complete the integration.

Advanced Intents in Dialogflow

Bevatel's integration with Dialogflow supports advanced intents, including agent handoff and interactive messages.

1. Creating a Handoff Intent

1. Create a new intent in Dialogflow named Handoff Intent.
2. Add training phrases such as "Talk to an agent" or "Speak with an agent."
3. Add a custom payload in the response section with the following JSON:

{
 "action": "handoff"
}

This payload triggers a handoff to a live agent in Bevatel when the customer requests to speak with one.

2. Creating an Interactive Message Intent

Bevatel supports interactive messages for the website channel. You can create custom payloads for options or cards. Here's an example for an options interactive message:

json

{
 "content_type": "input_select",
 "content": "Select your favorite food from below",
 "content_attributes": {
   "items": [
     {
       "value": "I like sushi",
       "title": "Sushi"
     },
     {
       "title": "Biryani",
       "value": "I like biryani"
     },
     {
       "title": "Pizza",
       "value": "I like pizza"
     }
   ]
 },
 "private": false
}

When the user interacts with the message, the selected option is sent back to Dialogflow, allowing you to configure follow-up intents based on the selection.

FAQ

Q: How can an agent transfer the conversation back to the Dialogflow bot?
When the bot is connected to an inbox, conversations initially have a pending status, allowing the bot to handle them. If an agent wants to push the conversation back to the bot, they can change the conversation status back to pending.

Q: What if the JSON key is invalid during setup?
Ensure that the key file is copied and pasted in its entirety, and that it matches the service account created for the Dialogflow API.