Defining suggested responses for a conversational channel

To automate business processes and save the time that case workers spend answering simple questions, you can define suggested responses so that Pega Intelligent Virtual Assistant (IVA) reacts to user input correctly. The IVA detects the best response for a topic using text analysis and automatically invokes its response type. For example, when a user asks for help during an interaction with the IVA, the system can display a text message asking for more information and a menu of available options.
Before you begin: Add suggested cases to the IVA. For more information, see Adding suggested cases for a conversational channel.

You can also add suggested cases to the IVA, so that the system automatically starts a case in the application based on user input. All suggested cases and suggested responses that you define are topics.

  1. In Dev Studio, click the name of your application, and then click Channels and interfaces.
  2. In the Current channel interfaces section, click the icon that represents your existing Unified Messaging or Web Chatbot channel.
  3. In the channel, click the Configuration tab.
  4. In the Content section, select a method for modifying a suggested response:
    • To add a new suggested response, click Add response.
    • To edit an existing suggested response, click the Switch to edit mode icon for a response displayed in the list.
  5. In the Response type list in the Response configuration window, select and configure a response type:
    Choices Actions
    Authenticate the user in the system
    1. Select Authenticate.
    2. In the Response text area, enter the text that you want to display.
    Cancel the current work in the interaction session
    1. Select Cancel.
    2. In the Response text area, enter the text that you want to display.
    3. Optional: To add a command to a menu you display after you cancel current work, click Add response.
    4. In the Menu item list, select the command that you want to invoke for the menu item.
    5. In the Label field, enter the menu item name that you want to display in the system.
    6. To add more commands that you want to display as menu items, repeat steps 5.c through 5.e.
    Display a rich text menu by using a data transform in the IVA for Web Chatbot
    1. Select Custom.
    2. In the Data transform field, select a data transform rule.
      The rich text menu in the IVA for Web Chatbot looks similar to a Facebook generic template. The menu consists of a title, subtitle, link to an image, and up to three menu options. You can use the default pyPopulateOutboundData data transform as an example for the IVA.
    Link to a specified Facebook account in the IVA for Facebook
    1. Select Link.
    2. In the Response text area, enter the text that you want to display to the user.
    3. In the Title field, enter a title for the Facebook account to which you want to link.
    4. In the Authentication callback URL field, enter the authentication URL for the Facebook account to which you want to link.
    Display a menu with a list of available commands
    1. Select Menu.
    2. In the Response text area, enter the text that you want to display to the user.
    3. Optional: Add a command to a menu by clicking Add response.
    4. In the Menu item list, select the command that you want to invoke for the menu item.
    5. In the Label field, enter the menu item name that you want to display in the system.
    6. To add more commands to display as a menu item, repeat steps 5.c through 5.e.
    Restart work on a case that is currently being processed in the session
    1. Select Restart.
    2. In the Response text area, enter the text that you want to display to the user.
    Display a text message to the user
    1. Select Text.
    2. In the Response text area, enter the text that you want to display to the user.
  6. Optional: To save the text from the Response text area in a template, click Add existing.
  7. In the Response command field, enter or modify a user command for the response type.
    Through text analysis, a user command provides the input that determines the best suggested response type that the system detects.
  8. If the suggested response requires authentication, select the Requires authentication check box.
  9. Optional: To add more suggested responses, repeat steps 4 through 8.
  10. Click Submit.
  11. Click Save.
What to do next: Configure text analyzer settings for the IVA. For more information, see Configuring text analyzer settings.