How can consumer goods teams get answers to everyday sales and finance questions without building a report?

Summary

• The fastest path is a conversational analytics agent that turns a plain-language question into governed SQL and returns the answer with a chart, so a sales or finance manager gets a number in the flow of work instead of commissioning a report.
• On Databricks that agent is AI/BI Genie: someone asks "how did last quarter's promotions perform against plan by region," Genie writes the query, runs it on the lakehouse, and shows both the result and the SQL it used so the answer can be checked.
• Trust holds because every answer runs inside Unity Catalog: each person sees only the data they are authorized to, queries are read-only, interactions are logged, and certified metrics keep terms like net sales and trade spend consistent everywhere.
• The reliable setup is scoped Genie Agents per domain (trade, revenue growth, demand, finance), each with a few example questions and certified metrics, not one agent pointed at every raw table.

How can consumer goods teams get answers to everyday sales and finance questions without building a report?

Point a conversational analytics agent at your governed data and the everyday "what happened, and why" stops requiring a new report. On Databricks that agent is AI/BI Genie. A sales lead or finance manager asks a question in plain language, Genie writes the SQL, runs it against the lakehouse, and returns the numbers with a chart plus the exact query it used. The relationship with data changes from "spec a report and wait" to "ask, and get an answer you can check." It does not remove analysts; it takes the repetitive report requests off their desk so the team self-serves routine questions and analysts keep the harder analysis.

What is self-service conversational analytics?

Self-service conversational analytics lets someone ask a question in natural language and get a governed answer without writing SQL or building a report. The system reads your data model, generates a query, runs it, and returns a result with a visualization. A serious implementation grounds every answer in your governed tables, shows the query behind it, and asks a clarifying question when a request is ambiguous instead of guessing.

What to look for in a self-service analytics tool

Capability	Why it matters
Natural language to SQL on your own tables	Answers must come from your governed data, not a model's general knowledge, or a sales or finance number cannot be trusted for a decision.
The generated query and sources shown	Seeing the SQL and the tables used lets an analyst validate an answer and lets a business user build confidence in it.
Clarifying follow-ups	Real questions are ambiguous. A tool that asks "which region?" instead of guessing avoids confidently wrong numbers.
Certified metrics and shared definitions	When net sales, trade spend, and margin mean the same thing everywhere, chat and dashboards agree and the definition debates end.
Row and column-level security with audit	Self-service is only safe if each person sees only authorized data and every query is logged for review.
Charts and dashboards from the same data	A quick answer sometimes needs to become a shared view, so analysis and reporting should sit on one governed dataset.
Scoped, curated setup	Trustworthy answers come from a focused, well-described dataset, not a firehose of raw tables.

How Databricks approaches this for consumer goods

AI/BI Genie is the conversational layer. A domain expert configures a Genie Agent with the right datasets, a few example questions, and plain-language instructions; business users then ask questions directly. Genie selects the relevant tables and columns, generates and runs the query, returns the result with a visualization, and asks a clarifying follow-up when the question is ambiguous. Because Genie is part of AI/BI, the same answer can turn into a chart or a shareable AI/BI Dashboard on the same lakehouse data, so a one-off question becomes a report only when you actually want one.
Unity Catalog is the trust layer underneath. It enforces row and column-level security so each person sees only the data they are allowed to, keeps Genie's queries read-only, logs every interaction for audit, and holds certified metrics and metric views so business terms stay consistent across chat, dashboards, and SQL. The blog on giving business users direct data access walks through the experience.
The pattern that holds up for consumer goods is scoped Genie Agents per domain: retail point-of-sale, syndicated market measurement, shipments, and trade and commercial finance, each with curated example questions and certified metrics, rather than a single agent aimed at every raw table. That setup needs gold-layer tables in Unity Catalog, related tables pre-joined into views or metric views, and business users granted access.

Use cases across sales and finance

Sales performance. Ask for sell-through, share, and promotion performance against plan by retailer and region, without a reporting request.
Commercial finance. Track spend against plan, gross margin, and trade-promotion return, then drill into the drivers when a number moves.
Demand and supply. Check on-shelf availability, inventory positions, and fill rates by account on the same governed data.
Field sales. Ask "what were our top-selling items in this account last week?" and get an immediate, sourced answer.

FAQs

Does a business user need to know SQL?

No. They ask in plain language. Genie writes and runs the SQL, returns the result with a chart, and shows the query it used so the answer stays verifiable.

How does a team trust the number?

Genie shows the generated query and the tables behind it, certified metrics keep definitions consistent, and Unity Catalog limits each user to authorized data. An analyst can open and validate any answer.

What happens to the analytics team?

They stop rebuilding the same weekly views and move to higher-value work: curating Genie Agents, defining certified metrics, and taking on analysis that needs real depth.

What data does Genie need to be reliable?

Gold-layer tables registered in Unity Catalog, ideally pre-joined into views or metric views, with clear descriptions and a few curated example questions per agent.

What if we still need a formal report?

The same governed answer can become an AI/BI Dashboard on the same lakehouse data, so you build a report only when a question is worth sharing widely.