What is a data mesh?
A data mesh is an organisational approach to analytics data that decentralises ownership, letting each business domain own and serve its data as a well-documented 'data product', rather than funnelling everything through one central data team. It emphasises domain ownership, self-serve infrastructure and federated governance.
Data mesh is a response to central data teams becoming bottlenecks. Instead of one team owning all pipelines and models, each domain, sales, finance, operations, owns its own data and publishes it as a reliable, documented data product for others to use.
Its four principles:
- Domain ownership teams own the data they know best.
- Data as a product treated with quality, docs and SLAs.
- Self-serve platform shared infrastructure so domains can build.
- Federated governance common standards, decentralised execution.
Data mesh is an operating model, not a specific tool.
iDBQuery fits a mesh well because it makes each domain's data directly and safely accessible without central pipeline work. Domain teams can connect their own sources and let their people ask plain-language questions with cited answers, true self-serve. When a question spans domains, iDBQuery's federation joins across those sources in one live model. And because it queries in place with role-based access and cited, auditable results, it supports the federated-governance side of the mesh: local ownership, consistent standards, and transparency into where every number came from.
Updated 2026-06-22