Two platforms at the core. Methods that fit the problem.
We work mainly with Snowflake and Microsoft Fabric. Which platform fits is decided by your existing systems, your team and your data types, not by our preference.
Snowflake or Microsoft Fabric?
The most common question in a first conversation. Both platforms support a data warehouse and AI applications; they differ mainly in how they fit your existing landscape.
| Criterion | Snowflake | Microsoft Fabric |
|---|---|---|
| Approach | SnowflakeStandalone cloud data platform, running on AWS, Microsoft Azure or Google Cloud. | Microsoft FabricIntegrated SaaS analytics platform in the Microsoft ecosystem, with OneLake as shared storage. |
| Building blocks | SnowflakeData warehouse, data engineering and the Snowflake Cortex AI functions, including Cortex Search, Cortex Analyst and Agents. | Microsoft FabricData Factory, data engineering, data warehouse, lakehouse, Real-Time Intelligence and Power BI in one environment. |
| A good fit when … | Snowflakea cloud-independent platform is wanted, or SQL- and dbt-based engineering is central. | Microsoft FabricMicrosoft 365, Azure and Power BI are already in place and the tools should come together in one environment. |
There is no universal winner. We recommend the platform based on your licences, team skills, data types and expected operating effort.
Vendor documentation: Snowflake · Microsoft Learn
Methods and tools
What the terms mean, and in which service we use them.
- Data Vault
A modelling approach for historized, traceable data from many sources. Not a platform, but a decision about the data model.
Used in Data warehouse- dbt
Transformations as versioned, tested code instead of manual scripts. Makes data models verifiable and maintainable.
Used in Data warehouse- Data lakehouse
Combines the low-cost storage of a data lake with the analysis capabilities of a warehouse; useful with lots of unstructured data.
Used in Data warehouse- RAG
Retrieval-augmented generation: a language model answers questions from your own documents and cites the source.
Used in RAG solutions- Snowflake Cortex
AI functions directly on the data in Snowflake, for example search across documents or questions about structured data.
Used in RAG solutions- AI agents
Applications that work through a task in several steps with defined tools and limits.
Used in AI agents
Facing the platform decision?
Discuss your data platform