Sentiment analysis for email interactions

  • Release version: Australia
  • Updated March 23, 2026
  • 3 minutes to read

    • Configure Sentiment analysis for email interactions to evaluate sentiment, trends, and reasoning at interaction level and help agents resolve cases more efficiently.

    Before you begin

    Role required: admin

    Install the plugin for email interactions.

    About this task

    Sentiment analysis for email interactions is generated from the information that you enter in the following fields:
    • Description
    • Short Description
    Any modifications to the names or labels of these fields can result in issues with sentiment analysis. Sentiment is calculated using several factors:
    • Latest requester message
    • Overall conversation between agent and customer captured in transcript
    • Sentiment history (trend)
    • Issue complexity
    • Feedback or continued frustration at the end of conversation
    • Frequency of updates and response time
    • Language tone
    The Sentiment Reasoning field in the record provides details on which factors were considered for reasoning. This includes:
    • Tone of the requester's messages across the transcript.
    • Feedback or continued frustration in the latest requester's messages.
    • Complexity of the issue based on interactions from both the requester and the agent.

    Procedure

    1. Navigate to Admin > Now Assist Admin > Skills.
    2. Select the Customer workflow, and CSM as the product.
    3. Activate skill for the Sentiment analysis for email interactions skill.
      Each skill has a guided setup with multiple steps. A check symbol next to each step indicates whether its setup is complete, partially complete, or incomplete. After configuring a step, select Save and continue to move forward, or Back to return to a previous step.
    4. Select General details and review the name and description of the skill.
      Additional information regarding details of the skill are displayed, but can't be edited.
    5. Select Choose Input and review the tables and fields to create prompts that determines where data is pulled from.
      Note:
      You cannot modify the input data source.
      Table 1. Table 1: Inputs for the Sentiment analysis for email interactions skill
      Input Description
      Input table Interaction
      Input fields
      • Description
      • Short Description
    6. Select Role attribution
      Note:
      You cannot modify the input data source in base system.
      Table 2. Table 2: Inputs for the Sentiment analysis for email interactions skill
      Input Description
      Input table Interaction
      Requester fields
      • Opened for
      • Created by
      Fulfiller fields Assigned to
    7. Select Define Availability to customize how and when the skill capability is active and accessible.
      • Select Skill is always available so no restrictions are placed on when a skill is available.
      • Select Customize skill availability to define conditions and use the condition builder to configure fields and values.
    8. Select Define Trigger and edit the job schedule frequency period to configure how often the skill will be triggered.
      To enable a scheduled job for a custom table, follow these steps:
      1. Navigate to All > System Definition > Scheduled Jobs.
      2. Search and select Sentiment analysis scheduler job interaction CSM.
      3. Replace the table name interaction with the name of your custom table in the script.
      4. Select Update.
    9. Select Save and continue.
    10. Select Define access to determine who can access this skill.

      By selecting specific roles, you're controlling who can use it. The roles you choose will also be available in the next step Select display.

      Default and Custom Roles:
      • If no changes are made, the default roles sn_customerservice_agent and sn_customerservice.consumer_agent will automatically appear in Define Access and Select Display.
      • If custom roles were added before the upgrade, they’ll be updated automatically by a script.
      • If new roles are created after the upgrade, you must manually add them in both the Define Access and Select Display.
        Note:
        In the Select Display step, you can only choose roles that were added in the Define Access step. If you add a role in Define Access, you still must manually select it in Select Display to make it active.
    11. Select Save and continue.
    12. Select display and toggle to determine if the skill appears in In-product desktop, displaying Sentiment analysis for email interactions in all CSM products.
    13. Select Save and continue.
    14. After selecting Review and Activate to examine changes, select Done to close the sentiment analysis generation settings.
    15. Select Activate to turn on the skill for agents and complete the configuration.

    What to do next

    • Now Assist Skill Kit is the ServiceNow framework for cloning and customizing Now Assist skills. When configuring Sentiment Analysis for email interactions, the base table provides the core email record data. Additional inputs supply the LLM with broader context- such as email thread history, linked case fields, or customer contact data that can improve the accuracy of sentiment output.
    • You can customize the email sentiment prompt. Follow the steps to customize:
      1. Access the OneExtend Capability Definition record by going to the sys_generative_ai_config table. Here is an example for Azure OpenAI (other providers like Claude and Gemini will have a different sys_id): https://anoway1.service-now.com/sys_generative_ai_config.do?sys_id=479cdabf93b48310003afba6dd03d606
      2. Go to the Prompt Template field and edit the prompt.
      3. Save the record.
        Note:
        Verify that the user is in Now Assist for Customer Service Management (CSM) scope and the record is still in Published state.