Data Mesh
Also known as: domain-oriented data architecture
Data Mesh is an organizational and architectural pattern that decentralizes ownership of analytical data to the domain teams that produce it, treating data as a product with explicit consumers, SLAs, and discoverability.
Detailed explanation
Data Mesh reacts against the monolithic data lake / central data team model that often becomes a bottleneck. It is built on four principles: domain-oriented ownership, data as a product, self-serve data platform, and federated computational governance.
In practice, each domain (orders, marketing, supply chain) owns the analytical datasets it produces, exposes them through standardized interfaces (often the lakehouse or warehouse), and treats internal data consumers as customers. A central platform team provides the substrate (storage, compute, catalog, governance) but does not own domain data.
Data Mesh is not the right answer for every organization — small teams may find it overhead-heavy. It works best when there are enough domains and data consumers that central coordination becomes a tax rather than an enabler.