Plain-language definitions of the terms used across the Data Product Standard. Where a term has a fuller explanation, its name links to it.

Core concepts

TermMeaning
Data productA curated, governed unit of data with a clear purpose, a named owner, and a defined way to consume it. Belongs to exactly one domain.
Data assetA single real piece of data in a source system — a table, file, dashboard, or API. Assets are the raw material a data product packages and publishes; on their own they carry no owner or contract.
DomainA business area, such as Sales or Finance, that owns a set of related data products. Every product belongs to exactly one.
Data MeshThe approach behind the model: data owned as a product by the domain closest to it, rather than by a central team.
Data MarketThe catalog where published data products are discovered and requested. Only products in production appear.

Ports and interfaces

TermMeaning
Output portThe interface through which a data product publishes data to consumers. Holds assets from one technology, governed by one contract.
Input portThe interface through which a data product reads from another product’s output port.
Management portAn operational interface for monitoring, health, and logging signals — kept separate from the data a product publishes.

Contracts, quality, and commercial terms

TermMeaning
Data contractThe agreement a data product makes about what it delivers through a port — schema, quality, service levels, and terms. Has its own lifecycle.
Service Level Agreement (SLA)The operational commitments a product makes to consumers — availability, freshness, and support response.
Data quality checkA rule that tests whether data meets a defined expectation, such as completeness, with a severity.
Pricing planHow access to a data product is priced, when it is not free.
LicenseThe terms under which a consumer may use a product’s data.
Data holderThe organization accountable for the data a product delivers.

Classification

TermMeaning
PII LevelClassification of how much personal information an asset contains; drives privacy obligations.
ConfidentialityClassification of who may see an asset; drives access rules.

Accountability

TermMeaning
OwnerThe single person accountable for a data product. Required before it reaches production.
StewardAccountable for a product’s meaning and quality; approves its data contract. Optional, not a gate.
CustodianRuns the pipelines and systems that produce the data.
Subject-matter expert (SME)Consulted for domain context; not accountable for delivery.
ProducerThe team or on-call group behind the product’s output.
ConsumerA team that uses a data product. Tracked through relations, not an ownership role.

Lifecycle and standards

TermMeaning
Workflow stateThe point a product has reached in its lifecycle, from announcement to retired.
Dawiso Unified Data Product ModelDawiso’s single model for a data product, combining ODPS for the product, ODCS for contracts, and Data Mesh ports. Every field traces back to a published standard; it is not a standard in its own right.
ODPS (Open Data Product Specification)The open specification Dawiso uses to describe the data product itself, from opendataproducts.org, version 4.1. Bitol publishes a different standard with the same acronym, which Dawiso does not use.
ODCS (Open Data Contract Standard)The open standard, from Bitol, that Dawiso uses to describe data contracts. Version 3.1.0.