Skip to main content


         This documentation site is for previous versions. Visit our new documentation site for current releases.      
 

Editing the conversation smart shapes

Updated on November 18, 2021

Ensure that Pega Intelligent Virtual Assistant™ (IVA) can respond correctly by asking questions, displaying text messages, or starting another embedded conversation flow by configuring smart shapes for a conversation process.

For example, for a conversation process that the system uses to dispute a cancellation fee for a requested vehicle ride, you can configure the Ask a question smart shape to ask users for the date when the fee was applied and use smart validation to ensure that the response is in the correct date format.

Instead of running in the context of the work item, you can also configure the conversation flow to run in an embedded page if you create a nested data model for your case using the object-oriented approach. For example, you can create a Change address case with two data objects of the same Address type, one specifying the customer's current address and another their registered home address. To ask users about both their current and registered home address, you can define and reuse the Update address conversation flow that runs in an embedded page first for the current address property, and then for the registered address property.

Note:
  • If your channel does not include a translation for the fields for a Send a message or Ask a question smart shape, the system will use placeholder text specified in the default language for those fields.
  • If you translate the text into another language for a Send a message or Ask a question smart shape, the system creates a new language-specific ruleset only if the ruleset does not already exist for the language. If the ruleset already exists and is locked, to translate your text, ensure that you first unlock the ruleset in your application.
  1. Create or edit a conversation process for a conversation channel.
    Result: The system displays a flow canvas with shapes for the conversation process.
  2. In the flow, right click a conversation smart shape for which you want to edit properties, and then click View properties:
    ChoicesActions
    Send a message shape
    1. If you configure multiple languages for the chatbot, in the Configure conversation window, select a language for the message text.
    2. In the Message text data field, enter the text of the message to display to the user.
    3. Optional: To configure additional languages for the message text, repeat steps 2.a through 2.b.
    4. Optional: To use text from an existing template, click Use existing.
    5. Click Submit.

    For example, in a conversation process related to a Fee Inquiry case, you can configure the shape to display information that a charged cancellation fee for a requested vehicle ride will be refunded within the next few days.

    Ask a question shape
    1. In the Configure conversation window, in the Question type list, select the type of question: Open ended, Buttons, Quick replies, or Yes or No.
    2. If you configure multiple languages for the chatbot, select a language to define for the message text, question option labels, and validation error message text.
    3. In the Message text data field, enter the text of the question to display to the user.
    4. Optional: To use text from an existing template, click Use existing.
    5. If you did not select an open-ended question type, in the Options column, enter question options in each row in the selected language, and in the Connectors column, select whether you want to use a connector for the option.
      The question options for which you do not select a connector automatically use the Else connector. For the yes or no question type, the system does not provide option connectors, only the read-only Yes and No connectors.
      For example: If you define the buttons or quick replies question type that the system uses to ask which color is the user's favorite, you can enter the following question options: Blue, Green, Yellow, Red, and Black.
    6. Optional: To add additional question options, click Add item and repeat step 2.e.
    7. Optional: To save a response to the question to a field, expand the Advanced section, and in the Save response to the field list, select a field in which to save the response.
    8. Optional: To extract an entity from an open-ended question, in the Entity to extract list, select an entity type.
      For example: To extract an answer that is in a date format, select Date.
      This field gives you the ability to find the entity in user answers and automatically validate that the answer is in the correct format, for example, date or email format.
    9. Optional: To repeat asking the question once, twice, three times or until a user enters a valid response, in the Number of repeats list, select 1 time, 2 times, 3 times, or Until successful.
      The chatbot uses smart validation and depending on the selected option for this step, the system continues to ask the question until a valid response is found. The system uses an additional connector if the chatbot does not find a response after asking the question the specified number of times.
      For example: You can use smart validation to ask for an answer in a date format a maximum of three times. If the user does not provide a date three times, then the chatbot uses the additional connector in the process to transfer the conversation to a customer service representative (CSR) so that they can assist the user further.
    10. In the Validation error message field, enter or edit the message in the selected language to display if the answer is not valid.
    11. Optional: To configure additional languages for the message text, question option labels, and validation error message text, repeat steps 2.b, 2.c, 2.e, and 2.j.
    12. Click Submit.

    For example, in a conversation process related to a Fee Inquiry case, you can configure the shape to ask the reason for disputing the cancellation fee using the Buttons question type with three options specifying that ride did never arrive, the ride was late, and another reason. The system saves the user's response to the multiple-choice question in the Dispute Reason field.

    Conversation flow shape
    1. In the Configure flow window, in the Define flow list, select On current page, or On embedded page and then in the Select property list, select a property and a flow.
      To use the conversation flow in the context of the current work item, select On current page. If you use an object-oriented approach to the case data model that contains a nested structure, select On embedded page, and then select the context of which nested property to run the conversation flow.
    2. In the Select conversation flow list, select the name of the conversation flow to use within the conversation process.
    3. Click Submit.
    For example: The following figure shows the properties for the Ask a question shape that the system uses in a conversation process designed to dispute a cancellation fee for a requested vehicle. With this smart shape the chatbot can ask users for the reason for disputing the cancellation fee by giving them three options displayed as buttons to chose from. If you configure multiple languages in your chatbot, you can define the message text and option labels in another language from English by first selecting that language from the list.
    The properties for the Ask a question smart shape in English
    The Configure conversation window where you can configure the properties for the Ask a question smart shape.
    For more information about flow shapes, see Flow shapes.
  3. Click Save.
What to do next: Test your configured conversation shapes in the conversation process using the preview console. For more information, see Simulating a conversation and building a chatbot.
  • Previous topic Adding a stand-alone conversation process for a conversational channel
  • Next topic Adding a case type conversation process for a conversational channel

Have a question? Get answers now.

Visit the Support Center to ask questions, engage in discussions, share ideas, and help others.

Did you find this content helpful?

Want to help us improve this content?

We'd prefer it if you saw us at our best.

Pega.com is not optimized for Internet Explorer. For the optimal experience, please use:

Close Deprecation Notice
Contact us