Skip to main content
Verify the version tags to ensure you are consuming the intended content or, complete the latest version.

Configuring Pega GenAI suggested replies

Suggested replies are the automated and relevant responses that customer service representatives (CSRs) receive from Pega GenAI™ and Dialogs for customer inquiries during a live interaction with a customer. Pega GenAI uses underlying Pega Knowledge™ articles, common phrases, and customer account information as vetted sources of information for suggesting responses. This feature reduces the handling time and time to respond because CSRs do not have to browse and read knowledge articles or search for common phrases to provide the customer with an accurate answer.

Video

Transcript

This video shows you how to configure Pega GenAI suggested replies.

Consider a scenario in which U+, a retail bank, has integrated Pega GenAI and Dialogs into its existing customer service platform. Now, the bank aims to configure Pega GenAI suggested replies to analyze customer inquiries in real time and provide automated responses to CSRs during live interactions. This integration eliminates the need for CSRs to search for information or browse through knowledge articles manually.

Note: You must configure Pega GenAI Knowledge Buddy to get relevant responses from Pega Knowledge to answer the questions. For relevant training materials, see Pega Knowledge Buddy.

To implement the business scenario, in the navigation pane of App Studio, click Control Panel > Engagement Channels > Messaging > Suggestions & AI features > Start.

In the Case "dialogs" for suggested replies section, select the Show dialogs as suggested replies checkbox to display the dialog as a suggested reply to the CSR.

Note: If you do not enable this option, the system receives the suggested reply from the underlying Pega GenAI engine.
Suggestions and AI features

In the Pega GenAI suggestions section, select the Enable Pega GenAI suggested replies checkbox to receive suggested replies from Pega GenAI.

Now select any one of the options to set the number of messages that the system can pass to Pega GenAI. You can either select All to send every single message to Pega GenAI or select Custom if you want to send the desired number of messages to Pega GenAI. It is all about tailoring it to your needs.

Moving on to the Inputs to Pega GenAI suggested replies section. Here, you can provide inputs to Pega GenAI suggested replies:

  • If you want Pega GenAI to suggest replies based on knowledge articles, select the Knowledge articles checkbox.
  • If you want Pega GenAI to draw suggested replies from common phrases to help your CSRs, select the Common phrases checkbox.
  • If you want Pega GenAI to tailor its suggestions based on specific customer account details, select the Customer account information checkbox.

There is also a sample prompt outlined, but you can alter the prompts to fit their communication style better and tweak the resulting suggestions.

Save your changes.
Now, test the scenario with a live agent. A CSR logs into the Interaction Portal to address customer inquiries. A customer initiates a chat session with a live CSR, reporting no audio when using a new cable.

The CSR receives an incoming call from a customer, accepts the call, and then asks for the account number or Email address of the customer.

The customer enters her name and Email address in the chat window. The CSR searches for the Email address and finds the account of the customer, and then asks how they can help.

Mrs. Biggs states that she is facing a cable issue.

The CSR can click Suggested Reply to generate a reply from Pega GenAI. He can then expand the widget to see the suggested reply received from Pega GenAI for the issue mentioned by the customer.

The widget also displays error messages if something goes wrong. In the widget, the CSR can either send the reply to the customer or move it to the compose area for editing prior to sending.

You have reached the end of this video. You have learned:

  • How to configure Pega GenAI suggested replies.

This Topic is available in the following Module:

If you are having problems with your training, please review the Pega Academy Support FAQs.

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 Academy has detected you are using a browser which may prevent you from experiencing the site as intended. To improve your experience, please update your browser.

Close Deprecation Notice