Semantic layers are back on the agenda, and it’s all about AI. If you want a language model to provide accurate answers about revenue, margin, or churn, you first need to define what those terms mean. But which approach should you choose?
Joris Jonckers, Data Engineer at JUVO, compares the options based on his daily experience with Snowflake and dbt.
1. What a semantic layer is, what it delivers, and what it costs
2. A comparison of Snowflake Semantic Views, dbt (MetricFlow and the dbt Semantic Layer), and the Power BI semantic model
3. Reference architectures for each option and how Tableau and Power BI fit in
4. AI on the semantic layer: Snowflake Cortex and Power BI Copilot, and what’s needed for reliable answers
5. Pitfalls and Q&A
No sales pitch and no one-size-fits-all solution—because there isn’t one for every project. Instead, an honest assessment of where each option excels and where it falls short. The field is evolving rapidly, so consider this the state of affairs as of November 19 at 12:00 PM.