From Tables to Meaning: Building True Data Products with Ontologies
Every enterprise wants to be data-driven - and one of the best ways to achieve that is by treating data as a product. But simply taking the tables in your data warehouse and tying a ribbon around them doesn’t make them data products.
To productise your data, you need to make it appealing to your customers - the business users who rely on it. That means presenting the product in their language.
The truth is, most data products don’t speak the language of the business.
They speak in tables. In columns. In metadata and in schemas.
Meanwhile, the business talks about customers, risk exposure, margin erosion, and retention strategies. Somewhere between those two worlds - technical schemas and business meaning - most of the value is lost.
⚡ That’s not a data problem. It’s a meaning problem. ⚡
Enter the business ontology - a formal, machine-readable way of defining how your business actually thinks. Ontologies don’t just describe data structures; they define concepts like “Active Customer,” “Quarterly Recurring Revenue,” or “Supply Chain Disruption,” and map how those concepts relate to one another.
The simple, logical step then, is to make each data product align with a concept in the business ontology. The Supply Chain product should give you all the Disruptions. The Customer product should give you the Active Customers.
That’s the idea behind DPROD - an open Data Product Ontology that allows you to describe your data products in a way that connects each product’s output to a concept in your ontology.
To connect your data to your meaning.
This is a powerful shift. Users can now discover the data product they need by searching through the concepts in your ontology. Your business meaning becomes the entry point to your entire data landscape.
Think about it: each new data product becomes a building block in your Semantic Layer - a decentralised, self-organising map of business meaning that grows with every product you deliver. Your ontology gives structure to the data; your data gives life to the ontology.
Plug your LLM agents into this setup, and you’re on a path to making your organisation truly ready for the age of AI.
No moonshot re-platforming required. This is a pattern that scales organically - one use case, one data product, one concept, at a time.
⭕ DPROD: https://ekgf.github.io/dprod/
⭕ Semantic Layer: https://www.knowledge-graph-guys.com/blog/the-semantic-layer
⭕ Meaning Is Your Mote: https://www.knowledge-graph-guys.com/blog/semantics-is-meaning-the-hidden-structure-that-makes-you-unique
