Virtual Agent chat widget interface

  • Release version: Xanadu
  • Updated March 5, 2025
  • 3 minutes to read
    • Summarize
    Summarized using AI
    This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.

    Summary of Virtual Agent chat widget interface

    The Virtual Agent chat widget interface enables automated, conversational interactions between users and the Natural Language Understanding (NLU) Virtual Agent within ServiceNow’s service portals, including CSM, HR Service Delivery, and ITSM. It provides a user-friendly messaging environment where users can select topics, ask questions, request live agent support, send attachments, and receive audio/visual notifications. This interface is based on the Service Portal web widget and is specifically designed for NLU Virtual Agent functionality.

    Show full answer Show less

    Key Features

    • Topic Selection and Interaction: Users can start conversations by typing requests or browsing a searchable list of all available topics. The Virtual Agent guides users through prompts, choices, and rich controls like buttons or image carousels to assist in completing tasks.
    • Live Agent Transfer: Users can seamlessly transfer from the Virtual Agent to a live support agent. The chat window updates to reflect the live interaction and enables sending of attachments (images, text, PDFs) to the agent.
    • Attachment Handling: File uploads during live chat sessions are virus scanned automatically. If a scan fails or the upload times out, users receive clear messages to retry. Note that file uploads are supported only in bot conversations on mobile, not during live agent chats.
    • Notifications and Alerts: Users receive visual indicators of new messages and can enable or disable audio alerts via toggle switches in the chat menu. The default alert sound can be customized or disabled through system parameters.
    • Small Talk Recognition: The chat widget detects common conversational intents such as greetings, gratitude, complaints, and closure statements using the Semantic Filter Framework. Complaints can automatically prompt escalation options like live agent transfer or support request creation.

    Practical Benefits for ServiceNow Customers

    • Improves user experience by providing an intuitive, interactive chat interface for self-service and support.
    • Reduces resolution times by enabling instant topic discovery, automated assistance, and smooth escalation to live agents.
    • Enhances communication flexibility with support for attachments and customizable notification settings.
    • Supports multilingual small talk handling to maintain natural, conversational engagement.
    • Ensures security and compliance through automatic virus scanning of uploaded files.

    The Natural Language Understanding (NLU) Virtual Agent chat widget interface captures an automated messaging session between Virtual Agent and the user. Users can see a list of all the topics that the bot can help with, enter a request, request a transfer to a live agent, send and receive attachments, and receive visual and/or audio notifications.

    The chat widget interface uses the Service Portal web widget, which you can configure for custom service portals. This widget is available in the CSM, HR Service Delivery, and ITSM Service Portals.

    Note:
    The information in this content is applicable to only Natural Language Understanding (NLU) Virtual Agent. For more information about the LLM Virtual Agent end-user experience, see Using Now Assist in Virtual Agent.
    Figure 1. Simple conversation example — chat widget
    The chat widget window has icons to view support options and start a new conversation. Conversation elements include a welcome message, user and bot responses, and a closing message.
    A simple conversation typically includes the following elements:
    • Default welcome message from the virtual agent, followed by the first prompt for the conversation. The initial prompt asks the user to type a question or to choose from a list of available topics.
    • User response to the first prompt. In this example, the user chooses the topic for checking an IT ticket.
    • Series of prompts, acknowledgments, and responses exchanged between the virtual agent and the user until the users reach their goals or complete their tasks. You can use different rich controls in bot prompts, for example, choice lists with buttons or a carousel of images to select an item.
    • Closing message to end the conversation.

    When your users start a conversation with the bot, they can enter a request or see a list of everything that the bot can help with. If they choose to see everything, the chat window displays all topics available to the user. Your users can use the search box to filter the list of topics.

    Figure 2. See all topics
    List of All Topics displayed after user chooses to see everything the bot can help with.

    The conversational interface offers your users several options to manage the conversation. Users can stop the current conversation and start a new one, contact support to access a live agent for immediate assistance, and control audible alerts for chats.

    Figure 3. Support options menu
    The Support options menu lets users talk to a live agent if available, call or email support, and change notification settings.

    When users transfer to a live agent, the chat window header changes to indicate that they are now interacting with a live agent. The attachment and send icons are available in the chat window for uploading an image, text, or PDF file and sending it to the agent.

    Figure 4. Live agent transfer
    The chat window header no longer says "NOW Support." It now says, "LIVE AGENT."
    During live chat, users can attach a file of any type (for example an image, text, or PDF file) and send it to the agent. The system automatically performs a virus scan on the file attachment. If the scan is successful, the file is sent to the agent.
    • If the scan fails, a message informs the user that the scan did not pass so that the user can correct the file and then upload the file again.
    • If the upload and virus scan process times out (the upload takes too long), a message tells the user that the file couldn't be processed and to try the file upload again.
    Note:
    Most chat clients support file attachment uploads. However, the mobile chat client only supports file attachment uploads in bot conversations. Users cannot upload files in mobile chat with a live agent.
    Your users automatically receive audio and visual alerts when they receive a message from a live agent or virtual agent.
    • The visual indicator displays the number of messages received, next to the chat icon in the Service Portal window. When the user opens the chat, the indicator number closes.
      Figure 5. New message alerts
      New message alerts window.
    • Your authenticated end users can enable or disable audible alerts of chats through the toggle switch on the chat menu (Contact Support).
    • Authenticated end users can also enable or disable the delivery of notifications through the toggle switch on the chat menu (Contact Support).
    • The default sound notification uses the sn_va_web_client_alert.mp3 audio file.
      • You can change the sound that requesters hear by uploading your own audio file. For details, see Manage audio files.
      • To turn off the audio notification for your instance, use the system parameter sysparm_disable_audio_notifications=true. For example: https://<your-instance name>.service-now.com/nav_to.do?sysparm_disable_audio_notifications=true

    Small talk

    The chat widget recognizes small talk topics and small talk utterances from the Semantic Filter Framework. Small talk is supported by your primary and fallback languages. The semantic filtering recognizes these types of small talk:

    • Greetings: For example, Hi, how are you today?
    • Gratitude: For example, Thanks, this was helpful!
    • Complaint: For example, This answer doesn't help me.

      Chat conversations with complaints may prompt you to request a live agent or another fallback topic, for example, a create a support request topic.

    • Closure: For example, That's all I needed. Bye.