USE CASES : BUSINESS INTELLIGENCE
Business intelligence: one set of metrics for every BI tool.
Power BI, Excel, Tableau and Looker all query the same semantic model, so every dashboard shows the same number.
Revenue often gets defined in several places: a Power BI model, a Tableau extract, a Looker file and a few spreadsheets. AtScale defines each metric once and serves it live to every BI tool. ACE computes it in your data platform, so the same question gets the same answer wherever it’s asked.
The challenge
Each BI tool keeps its own copy of the metric logic.
Extracts and cubes go stale and cost money to refresh.
Dashboards built on raw tables can’t keep up at enterprise scale.
What you get
Every BI tool and AI assistant reads the same certified definitions.
Live queries stay fast because ACE answers from aggregates and cache.
THE CHALLENGE
When each tool defines revenue, the numbers don’t match.
Each BI tool comes with its own way to model metrics, and each team builds its own. Finance’s Power BI model and sales’s Tableau workbook calculate the same metric differently, and that’s how the board deck ends up with two numbers. Teams copy data into extracts to make dashboards fast, and then the copies drift.
HOW IT WORKS
Every tool queries the same model live.
| Tool | How it connects |
|---|---|
| Excel and Power BI | Live over XMLA with MDX and DAX, with no extracts and no drivers to install |
| Tableau and Looker | SQL over JDBC and ODBC |
| Google Sheets, Qlik, ThoughtSpot, MicroStrategy, Superset, Klipfolio, Hex | Supported connectors |
| Notebooks and apps | Python and REST |
| AI assistants and agents | The AtScale MCP server |
AtScale ACE turns each request into SQL for your platform and computes it where the data lives. It builds aggregates from the questions people actually ask and caches repeat answers, so dashboards stay fast and you don’t need extracts. Hierarchies, drill paths, time intelligence and semi-additive measures all come from the model, so every tool gets them the same way.
MOVING OFF LEGACY OLAP
A path off SSAS and other cubes.
Teams on SQL Server Analysis Services or other OLAP cubes can move their models to AtScale and keep the multidimensional features their users rely on, while the data moves to a cloud platform such as Snowflake, Databricks or BigQuery. Excel and Power BI users don’t have to change how they work.
THE PROOF
Some of the largest BI deployments run on AtScale.
AtScale has run in production for more than a decade at companies including a leading home improvement retailer, Vodafone, TELUS, Wayfair and Papa Johns. Snowflake embedded AtScale ACE in the Semantic Views XMLA endpoint, so Power BI and Excel can query Snowflake Semantic Views live.
FAQ
Questions about business intelligence.
Do we have to replace our BI tools?
No. AtScale serves the BI tools you’re already using, and they keep their own dashboards and reports.
Which BI tools does AtScale support?
Power BI and Excel connect live over XMLA, Tableau and Looker over SQL, Google Sheets through the AtScale Connector add-on, and Python notebooks through AI-Link. MicroStrategy is in beta and Qlik is experimental.
Do we still need extracts?
No. Tools query AtScale live, and ACE, the AI Computation Engine, answers from aggregates and cache so queries stay fast.
Which data platforms does AtScale support?
Snowflake, Databricks, Google BigQuery, Amazon Redshift, PostgreSQL and InterSystems IRIS. Self-hosted AtScale also supports Microsoft SQL Server, Azure SQL, Azure Synapse, Cloudera CDP and IBM Db2.
Where does the data live?
In your data platform. ACE computes there, and AtScale doesn’t keep a copy.
Do the same metrics work for AI?
Yes. AI assistants and agents query the same semantic model through the MCP server, so there’s no gap between their numbers and your dashboards.
Book a workshop.
The workshop takes thirty days from kickoff to results.
- Bring us your most controversial metric and the business questions your team and your agents ask about it every week.
- We benchmark your AI on them, with AtScale and without it.
- You keep the semantic model we built to run the test.