Use grounding to draft AI responses from a defined data source. Grounding indexes your objects and fields so that Einstein knows which information to base recommendations on. When you use grounding, your unique knowledge articles and case history add context and personalization to customer communications.
Required Editions
| View supported editions. |
Service AI Grounding Options
Service AI Grounding is supported for the listed Einstein for Service features only. For all other AI features, check for available grounding options in Salesforce Help.
| Feature | Available Objects | Description |
|---|---|---|
| Case Summaries | Case | Case fields, comments, and case feed are used to give Einstein details for drafting a Case summary. |
| Knowledge Creation | Case | Use the Case comments and emails to give Einstein details for drafting a Knowledge Article. See Map Einstein Knowledge Creation Responses to Knowledge Fields |
| Service Planner | Case | Choose the Case fields you want Agentforce to consider for service plans. See Set Up Service Planner Basics. |
| Service Replies for Email | Knowledge, Case | All email replies drafted with Einstein are grounded in your Knowledge base and Case context. See Service Replies for Email |
- Turn On Service AI Grounding
To get started, turn on grounding. After you enable Service AI Grounding, you can define the knowledge and case fields you want Einstein to ground on. - Ground with Knowledge
Select the title, summary, and additional Knowledge fields for Einstein to ground on. Einstein uses information from your selected knowledge fields to draft personalized responses. - Ground with Cases
Select the subject, description, and additional Case fields for Einstein to ground on. Einstein uses details from your selected case fields to draft personalized responses. To draft case summaries, you can also use case feed as a grounding source in addition to the selected case fields. Case feed includes crucial details like system updates, user comments, email messages, and status changes.