Glossary · 7 · Product information systems and data roles
Data product
Also known as: Data as a product
In data mesh, a data product is a dataset that a domain team offers to consumers as a product: with a named owner, documentation of its meaning and structure, quality commitments and an interface through which other teams and systems can find and use it.
- Data management
- Data governance
In one sentence
In data mesh, a dataset a domain team offers as a product, with an owner, documented meaning, quality commitments and a defined interface.
Example
The engineering team offers “released parts with revision and status” as a data product with a schema and a daily update promise; the CCMS and the portal subscribe to it instead of copying tables.
Explained in context
Context cards connect this term with others to answer one question. Also in British English and German.
How it applies
- Properties: Dehghani describes data products as discoverable, addressable, trustworthy, self-describing, interoperable and secure — each with an owner who commits to their quality.
- Documentation: Treating released product data as a data product gives documentation systems a defined, owned interface instead of an unnamed export; the product still delivers copies, so the system of record stays where the values are decided.
- Not only data mesh: The term is also used loosely for any packaged, reusable dataset. In the established sense it implies an owner and commitments to consumers.
Data product vs. dataset
A dataset is data. A data product adds the owner, the documentation, the quality commitment and the interface — often written down as a data contract.
External references
- Zhamak Dehghani, How to Move Beyond a Monolithic Data Lake to a Distributed Data Mesh (2019): https://martinfowler.com/articles/data-monolith-to-mesh.html
- Zhamak Dehghani, Data Mesh: Delivering Data-Driven Value at Scale (O’Reilly, 2022)