AdminAudit and usageUsage overview

Assistant usage dashboard (Deprecated)

View as Markdown

This dashboard and table is now deprecated and is only available if you set it up prior to June 2, 2026. If you have not yet set up Assistant usage, set up the AI usage dashboard instead. To move your existing Assistant usage data into the AI usage table, see Migrate Assistant usage data to AI usage. The usage dashboard will be removed after September 15, 2026.

The Assistant usage dashboard displays usage data about Sigma Assistant, including questions asked, users using Assistant, sources accessed by Assistant, and performance metrics like response time, tokens consumed, and number of LLM calls.

For example, if you want to evaluate changes that you make to the AI context for data models configured for Sigma Assistant or the configured AI provider, use the Assistant usage dashboard to review the resulting effects on usage and token consumption.

The dashboard contains the following tabs:

Tab nameDescription
QuestionsDetails about questions asked of Sigma Assistant.
UsersDetails about users of Sigma Assistant.
SourcesDetails about sources used to answer questions asked of Sigma Assistant.
PerformanceDetails about total response time for Sigma Assistant, for the LLM associated with the AI provider, and tokens consumed.

User and system requirements

Assistant usage table in connection browser

If you are granted access to the relevant schema, you can also explore the Assistant usage data in the relevant table in the connection browser.

Assistant usage is stored in a table called ASK_<ORGANIZATION_NAME> in the configured location.

Column nameDescription
TimestampTime when the question was asked.
User QuestionNatural language question asked.
User IdUnique user identifier for the user in Sigma.
Data ClassificationLLM classification of the data needed to answer the question.
Selected SourceSource used to answer the question.
Question IdUnique ID for the question.
User NameName of the user who asked the question.
AgentWhat agent was used to answer the question: source selection, docs, or analysis.
ImprovementN/A
Warehouse Runtime MsHow long the query ran to return the data for the question, in milliseconds.
Llm Runtime MsHow long the LLM ran to process the question and response, in milliseconds.
TeamsSigma teams that the user who asked the question is a member of.
ModelLLM model used. Should match the supported model for the configured AI provider.
ErrorIf present, the error message.
ResultDescription of the result or result type. For example, chart.
TokensTotal tokens consumed for the input (question) and output (response).
Call CountNumber of LLM calls required to answer the question.