Enterprises have spent years codifying trusted business logic in their Power BI semantic models, reports, and dashboards. These models represent thousands of hours of work from data teams and business analysts to define the KPIs, calculations, and relationships that guide day-to-day decisions.
But as organizations embrace Agentic BI, AI-driven analytics powered by Snowflake Cortex Analyst and natural language query (NLQ), they hit a frustrating roadblock: their existing semantics are not portable. To take advantage of these new AI capabilities, teams are often asked to redefine everything from scratch.
AtScale eliminates this barrier with the industry’s first end-to-end “roundtrip” solution.
Why Semantics Matter in Agentic BI
Most NLQ and text-to-SQL engines rely only on raw database schemas and metadata. These lack the rich business definitions already captured in BI tools.
When a CFO asks:
“What was our gross margin last quarter?”
An AI without semantic context may:
- Struggle to align the right tables
- Apply an incorrect calculation
- Misinterpret the business-approved definition
Power BI already contains the trusted definition. So do the dashboards that finance, sales, and operations leaders use every day. The challenge is making those semantics portable, usable outside of Power BI, so every AI agent, BI tool, or platform produces the same answer.
The AtScale Approach: Power BI to Semantic Modeling Language (SML)
AtScale’s new Power BI to SML Converter provides the only path to reuse the semantic definitions you already have, instead of rebuilding them from scratch.
This is the first example of an actual round-trip workflow:
- Convert Power BI semantics into open-source SML
- Convert SML to Snowflake Semantic Views
- Consume zero redefinition
Key Benefits of Power BI → SML Translation
- Measures & Calculations → Business KPIs like gross margin, net revenue, or customer churn are preserved in SML.
- Relationships → Dimensional models defined in Power BI are translated directly into SML, keeping business context intact.
- Business Terminology → Names, hierarchies, and definitions remain portable, ensuring cross-platform consistency.
Because SML is open source, organizations avoid lock-in and ensure compatibility with the next generation of BI and AI tools.
Extending Semantics to Snowflake Cortex Analyst
One of the most powerful applications of this round trip is with Snowflake Cortex Analyst, Snowflake’s generative AI assistant.
AtScale’s SML-to-Snowflake Semantic Views Translator converts the open SML model into Snowflake-native Semantic Views. This ensures Cortex Analyst interprets and answers questions using the same business definitions already trusted in Power BI.
That means:
- A finance user in Excel pivots sees the same gross margin calculation.
- A data scientist in Cortex gets the same number.
- A CFO asking Cortex Analyst in plain English receives the same trusted answer.
No duplication. No redefinition. No drift. Just round-trip semantics across your BI and AI stack.
Looking Ahead: Agentic BI Powered by Open Semantics
Agentic BI is more than NLQ. It’s about ensuring every question, across every tool, BI dashboard, or AI agent, returns the same trusted answer.
By bridging Power BI, SML, and Snowflake Cortex Analyst, AtScale makes this vision real. And because it’s built on open-source SML, customers can finally use the semantics they already have, without recreating them.
AtScale is also available directly in the Snowflake Marketplace, making it simple to get started and extend the power of a universal semantic layer into your Snowflake environment.
Explore AtScale on the Snowflake Marketplace and see how your existing Power BI models can become the foundation for your AI-driven analytics future.
SHARE
Case Study: Vodafone Portugal Modernizes Data Analytics