Recommendations in Recommended Actions

  • Release version: Australia
  • Updated March 12, 2026
  • 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 Recommendations in Recommended Actions

    Recommendations in ServiceNow’s Recommended Actions enable agents to receive helpful suggestions for taking specific actions to resolve issues. These recommendations can include guidance, decision trees, field value suggestions, or playbooks for step-by-step resolution. They enhance agent productivity by providing context-relevant, actionable insights directly within their workflow.

    Show full answer Show less

    Types of Recommendations

    • Guidance: Suggested actions or information an agent can use or share.
    • Decision Tree: A guided troubleshooting flow for diagnosing and resolving issues.
    • Field Recommendation: Suggested values to populate specific fields.
    • Playbooks: Step-by-step procedural guides to follow for resolution.

    Creating and Managing Recommendations

    Recommendations are created within rules using the New Recommendation screen, where you select the recommendation type. This selection auto-populates the Action type field on the Recommendation form.

    The Recommendation form captures key details such as:

    • Rule and Context: Read-only fields linking the recommendation to its context.
    • Name: Identifier for the recommendation.
    • Action type and Action: Define what the recommendation does; available actions depend on the selected type.
    • Launch mode: For playbooks, determines how the playbook opens (side panel or record view).
    • Resource Generator: Optional tool to supply relevant inputs or resources dynamically.
    • Recommendation Hint: Text displayed to explain why the recommendation is suggested, enhancing agent understanding.
    • Order and Active status: Controls recommendation display order and activation.
    • Action Inputs: Parameters required to execute the recommended action, which can be static or dynamically populated from context or resource generators.
    • Group: Allows grouping of recommendations to organize related suggestions together (used in Workforce Optimization for ITSM).

    Practical Application for ServiceNow Customers

    By leveraging recommendations, agents are guided towards appropriate actions with minimal manual search or guesswork. The flexibility to define multiple recommendation types and tailor inputs ensures that recommendations are both relevant and actionable.

    Resource generators enhance recommendations by dynamically providing necessary inputs, improving accuracy and efficiency. Playbooks can be launched seamlessly within the agent interface, facilitating consistent adherence to resolution procedures.

    Recommendations must be created within rules and saved before they are available to agents. The intuitive form fields and configurable hints help maintain clarity and usability for agents interacting with recommendations.

    A recommendation is a way to suggest a helpful action to an agent. A recommendation includes the action and any relevant resources and inputs.

    There are three types of recommendations that are provided to agents:
    • Guidance: An action that an agent can take or information that they can share.
    • Decision tree: A guided flow to troubleshoot an issue and provide a solution.
    • Field recommendation: A recommended value to use for a field.
    • Playbooks: A playbook to run for step-by-step resolution

    Creating a recommendation

    When you create a recommendation, you can select the type from the New Recommendation screen. This screen lists the available types.

    New Recommendation screen that presents available recommendation types.

    Selecting a type on the New Recommendation screen takes you to the Recommendation form. The selected type is added to the Action type field on the Recommendation form.

    Recommendation form

    The Recommendation form provides information about a recommendation, including the selected action and any inputs and resource generators that the action requires.

    When you create a recommendation, you select an action type and an action. The next steps depend on the action type that you select. For example, guidances and field recommendations require action inputs while decision trees don’t.

    When you create a recommendation, provide this information and save the record. After saving, you can see the Action inputs form section. Inputs are parameters that go into an action and enable agents to complete that action. You can select relevant inputs manually, from the context record, context inputs, or from the resource generator.

    You can access the recommendation records by navigating to All > Recommended Actions > Recommendations. New recommendations can be created only inside the rules. For more information, see Create a recommendation.

    Table 1. Recommendation form fields
    Field Description
    Rule The name of the rule for the context. This field is read-only.
    Context The name of the context that the Rule belongs to. This field is read-only.
    Name The name of the recommendation.
    Action type The type of recommendation:
    • Guidance
    • Decision tree
    • Field recommendation
    • Playbooks
    This field is automatically filled with the selection from the New Recommendation screen.
    Action The action to select for the recommendation. The available actions are determined by the selection in the Action type field.

    You can also create an action for the selected action type of guidance, decision tree, or field recommendation.

    Launch mode

    Defines how to open the playbook from Recommended Actions panel:

    • Launch and show in side panel: The playbook opens in the Suggested Actions tab of the Recommended Actions Contextual Side Panel.
    • Launch only: The playbook opens in record view.
    Note:
    This field appears when you select the Playbooks option in the Action Type field.
    Resource generator

    The resource generator to select from the Resource generators list. You can also create a resource generator.

    Guidances and field recommendations can use resource generators to find relevant resources for the action inputs.

    If you change the value in the Resource generator field, you must save the recommendation to use the resource generator.

    Recommendation hint Text that tells the agent why a recommendation is being suggested. Hints apply to action types of guidance and decision tree.

    The hint appears at the top of the action cards in the Recommended Actions side panel.

    Order The order of the recommendations inside the rule.
    Active When enabled, the recommendation is active.
    Action inputs Action types and actions require inputs. These inputs are the parameters that enable agents to complete an action. You can choose static values for action inputs, or you can use a resource generator to supply values.

    If a context input is available for the context, it appears in the pill picker of the fields in Action inputs section.

    Group Group created by a defined grouping strategy that displays relevant recommendations together.
    Note:
    You might need to configure the Recommendation form to display the Group field. This field is available by default for Workforce Optimization for ITSM users.
    Note:
    The Action type, Action, and Recommendation hint fields aren’t shown for the AI search resource generator. You can configure the Action type and Action fields, and mapping between AI search outputs and guidance actions inputs in the Search result mapping related list.