Short answerUnity Catalog Metrics define KPIs once so dashboards, Genie and agents calculate them the same way, and Domains group data and AI assets by business area. Together they make AI answers consistent and trusted. Start with the ten KPIs leaders ask about most, then expand.
Key takeaways
- Define each KPI once, use it everywhere.
- Domains organize assets by business area.
- Consistent metrics are the foundation for trusted Genie answers.
When a dashboard, a Genie answer and an AI agent each calculate revenue differently, trust disappears. People go back to spreadsheets.
Metrics: define once, use everywhere
Unity Catalog Metrics lets teams define reusable, governed KPIs. A dashboard, a Genie space and an agent can all use the same definition of revenue, margin or active customer.
Domains: organize by business
Unity Catalog Domains groups data and AI assets into business areas, such as finance or customer. This makes data easier to find, and it gives agents a clear, limited context to work in.
How to start
- List the ten KPIs your leadership reviews every month.
- Agree one definition for each with its business owner.
- Build each metric once in Unity Catalog and point dashboards and Genie to it.
- Group the underlying tables into two or three domains.
This is unglamorous work. It is also the work that makes every AI tool on top of your data more reliable.
About the author
NeuralShore Engineering, Databricks delivery team
Databricks architects, data engineers and AI engineers at NeuralShore, DLytica’s Databricks delivery company. Leadership comes from IBM data and AI architecture.