Data product
A managed data offering with consumers, a quality promise and interfaces. The dataset itself and the pipeline are not duplicated.
Bundle → Layer → Finding → Questions Filled
2 bundles · 3 layers · 4 findings · 9 questions
Product definition Why a data offering exists and who answers for it.
Purpose and owner
The product's purpose, owner and target consumers.
Product identity
A named product with purpose and accountable owner.
- What purpose does the data product serve, and for whom?
- Who is the accountable owner of the product?
Product versus dataset
Separating the product from the datasets it delivers.
- Which datasets and interfaces does the product deliver?
- Is the product referencing datasets rather than copying them?
Offering and consumers What is promised to whom, and how it is verified.
Offerings
Terms under which consumers receive the product.
Offering terms
Interfaces, access terms and quality promise per offering.
- Which interfaces and terms does each offering provide?
- How is the quality promise measured and verified?
- Do different consumers receive the product under different terms?
Consumer agreements
Agreements with consumers and their obligations.
Consumer agreement
A consumer's agreement and the uses it allows.
- Which consumers have agreements, and for which uses?
- What happens to consumers when the product changes or is retired?
Classifiers Filled
- Family
- Enterprise profiles
- Category
- Enterprise subject
- Entry kind
- subject
- Domain
- EnterpriseDAT
- Industry
- Cross-industry
- Tags
- EM-DAT-02W2subjectDataProductDataProductOfferingDataConsumerAgreement
What it is Filled
A managed data offering with consumers, a quality promise and interfaces. The dataset itself and the pipeline are not duplicated.
Why it exists Filled
A managed data offering with consumers, a quality promise and interfaces. The dataset itself and the pipeline are not duplicated.
Distinguishing features Filled
- A data product has an explicit purpose and owner; a table without them is not a product.
- Its quality promise is stated so that it can be verified.
- A data product references the datasets and pipelines it uses and does not copy them.
- One product can deliver several datasets and interfaces under different terms.
What robots and AI may and may not do Filled
Must not
- Declare any table a data product without a purpose, owner and promise.
- Copy the underlying dataset into the product so that data versions diverge.
- Report the quality promise as met without measuring it.
- Give a consumer access beyond the terms of their agreement.
Only with a human decision
- Retiring a data product that has active consumers.
- Changing the quality promise of an offering.
May
- Check whether a candidate product has a purpose, owner and verifiable promise.
- Report measured quality against the product's promise.
- List consumers and terms affected by a planned product change.
Moral aspects Filled
- Consumers rely on promised quality for decisions that affect people and money.
- Clear ownership is needed so that someone answers for errors in the product.
Who is affected
- Data consumers
- Product owners
- People described by the data
Owners Filled
Steward
The data owner or head of analytics who answers for data assets and their fitness for use.
Master systems
- Data catalogue
- Data warehouse
- BI platform
- Data orchestrator
Links to other meta-models Filled
neighbor
- EM-DAT-01
references
- WM-DAT-008 - conceptual-candidate
What else AI and robots need to interact with it Filled
Identity and identifiers required Filled
- Identified by data product identifier in the catalogue, with offerings by product and offering version.
Direct properties not applicable Not applicable
Not applicable
Enterprise record contour: physical properties belong to referenced world models.
Recognition optional Filled
- A data product record names a purpose, an owner, offerings and a quality promise.
- Often confused with a dataset, a table or an API endpoint.
Capabilities and actions required Filled
- Consumers can discover products and their terms in one place.
- Quality promises can be monitored against measurements.
Hazards and failure modes required Filled
- Products without owners that silently degrade.
- Copies of data that drift from the source dataset.
- Consumers left without notice when a product is retired.
Standards and interfaces required Filled
- W3C DCAT data services and datasets.
- W3C ODRL for expressing permitted and prohibited uses.
Context of use required Filled
- Used in organisations that manage data as products offered to internal or external consumers.
- Products containing personal data remain subject to data protection law.
Sources Filled
- W3C DCAT: dataset, distribution, catalog
- W3C PROV and Data Cube: provenance and observation
- SDMX and BI/data catalog practice: indicators, breakdowns, definitions and releases
- W3C Data Catalog Vocabulary (DCAT) Version 3 - Checked live (HTTP 200) on 2026-10-07.
- W3C Data on the Web Best Practices - Checked live (HTTP 200) on 2026-10-07.
- W3C Data on the Web Best Practices: Data Quality Vocabulary - Checked live (HTTP 200) on 2026-10-07.
- W3C The RDF Data Cube Vocabulary - Checked live (HTTP 200) on 2026-10-07.
Open questions
- When does a dataset become a product?
- What outcome is promised to the consumer?
- How to link multiple datasets and interfaces?
- Установить границу и решение reuse/extend/new по действующим спецификациям.
- Подтвердить semantic crosswalk, права и source mastership.
- Выбрать immutable refs; провести проверки fixtures до заявления о публикационной готовности.
Machine files
Provenance
enterprise research programme · published-partial
Built from: enterprise/models/em-dat-02/brief.json, ver-cy/world-models/card-supplements/em-dat-02.json
Still in Russian: suggested owner, blocking decisions, vercy candidates. Translate in research/enterprise/i18n/units.en.json.