.png)
The Data Radio Show - Bought to you by the Data Innovators Exchange
Join us weekly as we sit down and chat about the Data revolution and how to get involved with it, whether you're a seasoned pro at the forefront of change or someone new to the field.
We interview industry insiders, people in the field and experts across the world to bring you the latest advice, trends and changes to the field.
With dedicated content made for Data Professionals, at any level of expertise, you can keep abreast of the fast paced changing world of Data Management right here.
Join us in our Dedicated Skool Community and join the conversations at https://www.skool.com/data-management-innovators-4116/about
and make sure you sign up for the Data Pro Newsletter right here: https://www.datapro.news/subscribe
The Data Radio Show - Bought to you by the Data Innovators Exchange
How important is Data Modelling in an age of LLMs?
#️⃣ How do you build robust, scalable enterprise Data Warehouses without code? DSharp has the answer. Join Kim Johnsson as we discuss the power of data modelling, low-code/no-code and the use of AI in building robust, scalable data systems.
Sign up for the free newsletter for Data & AI Engineers here: https://www.datapro.news
KEY TAKEAWAYS:
1. Why a Model-First Approach Makes Data Warehousing Easier
Kim emphasises that focusing on a business-friendly, UML-like model (instead of jumping straight to tables, hubs, links, or satellites) dramatically reduces complexity. By describing data concepts (e.g., “Person,” “Organisation”) rather than technical structures, teams work more closely with business requirements and accelerate design decisions.
2. Automation and Reusability for Faster Time-to-Value
A recurring pain point in BI and data warehouse projects is repetitive manual work. Kim explains how automation tools—like DSH Studio—eliminate repetitive coding (SQL joins, Data Vault table creation, etc.). This reusability approach cuts down project timelines, enabling teams to deliver insights more rapidly.
3. Simplifying Ongoing Maintenance and Change Management
Because the solution generates all of the underlying structures automatically, changes are handled at the model level. If the business logic changes (e.g., a new attribute or different key), teams update the model and regenerate the warehouse. This top-down approach reduces the technical burden of managing large numbers of tables and relationships over time.
4. Integrating AI and Language Models into Data Modelling**
Kim highlights how generative AI (like ChatGPT) can quickly create initial data models by interpreting written business definitions. Though still maturing, these AI-driven features can give teams a head start, automate routine modelling tasks, and spark more productive discussions with stakeholders.
5. Key Takeaways for Data Leaders and Practitioners:
🫵🏼 Focus on Business Concepts: Modelling at a conceptual level encourages alignment with real business needs.
🫵🏼 Leverage Automation: Tools that auto-generate Data Vault structures free teams from repetitive coding, boosting productivity.
🫵🏼 Foster Collaboration: A shared model (versus purely technical artifacts) lets both business and IT speak a common language.
🫵🏼 Adapt to AI: Early experiments with LLMs show promising ways to speed up data modeling and design.
🫵🏼 Trust Through Iteration: As Kim notes, once teams see the automation “just works,” they confidently embrace a fully model-driven approach.
- Join the Data Innovators Exchange for free at https://www.skool.com/data-management-innovators-4116/about
- Sign up for the free Data Pro Newsletter at https://www.datapro.news/subscribe