Knowledge/RAG Quality Data and Metrics

Monitor knowledge retrieval quality scores for your Agentforce and generative AI solutions in production and sandbox environments. Use built-in dashboards and reports to track run-time performance, view trends, and identify areas to improve. Retrieval Augmented Generation (RAG) quality metrics help you identify problem patterns, conduct root cause analysis, and fine-tune your RAG-powered solutions.

Required Editions

Available in: Lightning Experience

Available in: Enterprise, Performance, Unlimited, and Developer Editions. Required add-on licenses vary by agent type.

About Knowledge/RAG Quality Data and Metrics
Learn about the metrics used to score retrievals. This feature collects data at run time and calculates quality scores for RAG-powered knowledge retrievals.
Set Up Knowledge/RAG Quality Data and Metrics
Turn on Knowledge/RAG Quality Data and Metrics to collect and store, in Data 360, quality scores and other data about knowledge retrievals. For orgs with only one dataspace, it’s turned on by default when Agentforce is turned on.
Data Model for Knowledge/RAG Quality Data and Metrics
Use the data in these DLOs/DMOs to build reports, dashboards, and queries that provide insights into knowledge retrieval events.
Gain Insights from Knowledge/RAG Quality Data and Metrics
To analyze Knowledge/RAG Quality Data and Metrics data, you can use pre-built dashboards and reports, or you can build your own. You can also run queries on the DMOs.
Implement Data Governance Permissions for Knowledge/RAG Quality Metrics and Data Objects
Knowledge/RAG Quality Metrics and Data can contain sensitive information, such as personally identifying information (PII), that you want to redact or hide from unauthorized users. Use Data Cloud Data Governance to protect sensitive data by setting up dynamic data masking and field-specific data exclusion policies. Grant access to the impacted Data 360 objects according to an access policy.

Knowledge/RAG Quality Data and Metrics

Required Editions

See Also