Comparison
Data meshvsData product
Data mesh
each domain team owns and publishes its own data products rather than throwing rows over the wall to a central data team.
An organisational approach that pushes ownership of analytical data to the teams that produce it, with a central platform providing the shared tooling and standards. It is a response to the central team becoming a bottleneck that understands none of the domains it models. It only works with genuine platform investment and enforced standards; without those it is a central bottleneck replaced by twelve incompatible ones.
Full entry →Data product
the table has an owner, a documented schema, a freshness commitment and a stated audience, like an API rather than a by-product.
Treating a dataset as something built for consumers, with a contract, documentation, quality guarantees and someone accountable. The shift is from 'here is the output of our pipeline' to 'here is a thing you may depend on'. It is the unit of ownership that makes decentralised data teams workable, and adopting the phrase without the freshness commitment or the owner is the common way it becomes vocabulary rather than practice.
Full entry →