CHECKLIST, CONDITION 9 OF 10, read 13 Nov 2026
Metric defined: 0 of 17 tools check it by rule, 16 leave it to a person
- The test
- A governed metric or semantic definition exists for each measure the agent will report.
- Why it matters
- When a metric has no single definition, the agent invents one in SQL. The result is a clean, wrong number that disagrees with the dashboard.
- How to check it
- Look for the metric in the semantic layer or glossary with one approved formula, filters and owner. Compare the agent answer with the dashboard figure.
- What failing looks like
- Net revenue has four formulas in four dashboards, and the agent picks the one with refunds left in.
What each use case needs
| Use case | Need | Why |
|---|---|---|
| Metric agent | required | A metric agent must answer from one governed definition or it will make its own. |
| Coding agent | recommended | A coding agent that changes model logic should keep it consistent with the governed metric definitions. |
| Write-back agent | not needed | Write-back moves records into fields and does not compute or report a metric. |
0 tools check metric defined by rule
- None on 13 Nov 2026.
16 tools leave metric defined to a person
- OvalEdge, rank 1, 64.5.
- Euno, rank 2, 62.5.
- DataHub, rank 3, 61.0.
- OpenMetadata, rank 4, 61.0.
- Atlan, rank 5, 59.0.
- Secoda, rank 6, 58.5.
- Databricks Unity Catalog, rank 7, 58.0.
- dbt, rank 8, 52.5.
- Alation, rank 9, 50.5.
- Ataccama, rank 10, 49.5.
- Qlik Talend Cloud, rank 11, 46.5.
- Snowflake Horizon, rank 13, 45.5.
- Precisely, rank 14, 43.0.
- Actian, rank 15, 39.0.
- Informatica, rank 16, 36.0.
- Collibra, rank 17, 36.0.
1 tools show none found on 13 Nov 2026
- Monte Carlo, rank 12, 46.0.
What the sources say
- Atlan, AI-ready data for talk-to-data: "It is ready when net_revenue has one approved definition as code, a certified table and a verified question-and-SQL pair."
- Atlan, AI readiness assessment for data estates: "Every business concept in the question needs one certified definition with a named owner, not a team alias, and a review date."
- dbt Labs, AWS and the dbt Semantic Layer for AI agent revenue accuracy: "Their starting point: an AI agent that writes its own SQL against production data is a liability."
Other conditions
Sources
- Atlan: AI-ready data for talk-to-data, read 13 Nov 2026
- Atlan: AI readiness assessment for data estates, read 13 Nov 2026
- dbt Labs: AWS and the dbt Semantic Layer for AI agent revenue accuracy, read 13 Nov 2026
Last reviewed 13 Nov 2026.