Catalog & Cocktails: The Honest, No-BS Data Podcast TAKEAWAY - Your data should speak your language with Säde Haveri
7 snips
Feb 26, 2026 A concise take on why metadata must bridge systems and people. They highlight the recurring gap where data fails to speak business language. Conversation covers why AI flounders without organizational context and how managing metadata creates shared understanding. Practical focus on modeling, prioritizing core data, and making semantics machine-readable for better governance.
AI Snips
Chapters
Transcript
Episode notes
Metadata Bridges Systems And People
- Metadata is the bridge between systems and people that gives data business meaning.
- Säde Haveri argues most data projects fail from lack of meaning, not technology, so semantics must be prioritized early.
Align Semantics At The Organizational Level
- Organizational-level artifacts like data products, ontologies, KPIs, and processes are where shared semantics must live.
- Säde emphasizes you can't force users to twist business logic to fit raw data; align models instead.
Start With Modeling And Identify Core Data
- Do the modeling steps you often skip: conceptual, logical, and physical models to capture meaning before implementation.
- Start by identifying master data, core KPIs, and operational objects that drive processes and dashboards.