Can a store manager just ask a question about yesterday's sales in plain English and get an answer?

Summary

• Yes. With Databricks AI/BI Genie, a business user such as a store manager can ask a question like "What were yesterday's sales?" in plain English and get an answer as a number, table, or chart, without writing SQL.
• Genie translates the natural-language question into a SQL query against your governed data and returns the result together with the query it ran, so the answer is transparent and checkable.
• Genie runs inside your existing Unity Catalog governance, so each person sees only the data they are permitted to see.
• Published AI/BI dashboards include an Ask Genie button, so a manager can go beyond the built-in charts and ask follow-up questions right where they are already looking at the data.
• Genie keeps conversation context so questions can be refined ("only the North region," "just for July"), and it asks clarifying questions instead of guessing when a request is ambiguous.

Can a store manager just ask a question about yesterday's sales in plain English and get an answer?

Yes. Databricks AI/BI Genie lets a business user ask questions in everyday language and get an answer back as a text summary, a table, or a visualization. A store manager can type something like "What were yesterday's sales?" and Genie converts that request into a SQL query against the underlying data, runs it, and returns the result along with the query it generated. No SQL, notebooks, or BI-tool expertise are required, and because Genie shows the query it ran, the answer stays transparent and verifiable.

Why Databricks AI/BI Genie makes this possible

• Plain-language questions over your real data. Genie uses a compound AI system to interpret a business question, select the relevant annotated tables and columns, and generate an equivalent SQL query. It responds with the generated query, a results table, and a visualization. See Genie concepts.
• Ask directly from a dashboard. Published AI/BI dashboards include an Ask Genie button by default. When you publish a dashboard, Databricks automatically generates a companion Genie Agent from the dashboard's datasets, so viewers can explore beyond the predefined charts, for example asking for sales by region for a specific product line, without leaving the dashboard.
• Grounded in your business meaning. Domain experts curate a Genie Agent with well-documented datasets, SQL expressions for business semantics, example queries, and plain-text instructions, so Genie understands terms the way your organization uses them and returns trustworthy answers. See best practices for curating a Genie Agent.
• Governed by Unity Catalog. Genie operates within your existing Unity Catalog governance model, so data access stays secure and consistent and each user only sees data they are authorized to access.
• Conversational and honest about uncertainty. Genie keeps context within a thread so a manager can refine a question step by step, and when a question is ambiguous it asks a clarifying follow-up rather than returning an incorrect result. See AI/BI concepts.

Getting started

• Read AI/BI Genie is now generally available for an overview of natural-language analytics.
• Follow Genie concepts and best practices for curating a Genie Agent to set up a space for your sales data.
• Publish an AI/BI dashboard so business users get an Ask Genie button in context.
• Confirm your Unity Catalog permissions so Genie answers respect the access controls you already have.

FAQs

Does the manager need to know SQL?

No. Genie turns a plain-English question into SQL for you, runs it, and returns a number, table, or chart. It also shows the query it generated so the answer can be reviewed.

How does Genie know what "sales" means in our business?

Domain experts curate a Genie Agent with documented datasets, business-semantic SQL expressions, example questions, and instructions, so Genie interprets your terms the way your organization defines them.

Will people see data they shouldn't?

No. Genie runs inside Unity Catalog governance, so every question respects the access controls already configured and each user sees only the data they are permitted to see.

Can they ask follow-up questions?

Yes. Genie maintains conversation context within a thread, so a manager can refine a question (for example, narrowing to a region or a month), and Genie asks a clarifying question when a request is ambiguous.