It treats data as a product, not only an asset description
Most metadata standards describe datasets. ODPS describes the product around the data, which is closer to how businesses buy, manage, and scale data products.
The Open Data Products standards family gives data products a shared structure. Each standard has a clear role. Together, they help humans, platforms, and AI agents understand how data products are defined, organized, connected, and interpreted.
Open Data Product Specification. Defines the data product, including metadata, access, quality, SLA, license, pricing, and strategy.
Read ODPS 4.1Open Data Product Catalogs. Organizes products into catalogs, portfolios, use cases, objectives, KPIs, and signals.
Read ODPC 1.0Open Data Product Graphs. Connects products to use cases, business objectives, KPIs, signals, and relationships.
Read ODPG 1.0Open Data Product Vocabulary. Provides the shared terms and semantics used across the standards family.
Read ODPV 1.0Open Data Product Recipes. Draft specification for repeatable workflow contracts, execution modes, provider profiles, gates, and review steps.
Read ODPR 1.0 DraftThe standards family is designed for practical data product operations, not just metadata documentation. It connects business intent, machine-readable structure, governance, automation, and agent-ready context.
Most metadata standards describe datasets. ODPS describes the product around the data, which is closer to how businesses buy, manage, and scale data products.
The family is modular instead of one large specification trying to do everything. Teams can use ODPS, ODPC, ODPG, ODPV, and draft ODPR where each one fits.
ODPS YAML stays the source of truth, while compact context formats, sidecars, SDK-generated artifacts, MCP-style access, and workflow contracts make the same knowledge usable by platforms and AI agents.
Many governance frameworks stop at policy. ODPS turns governance into structured, reviewable, machine-readable product information that supports governance by design.
Product ideas, requirements, risks, and technical details often live in separate tools. The ODPS family gives companies a path to convert that material into product specs, catalogs, graphs, reusable profiles, workflow contracts, and agent-ready context.
AI agents need structured context before they act: what the data product is, where it belongs, how it connects to business value, which terms mean what, and which rules apply.
Draft ODPR adds the workflow contract layer. It lets teams declare repeatable validation, generation, localization, review, release, and other agent-safe automation around ODPS, ODPC, ODPG, and ODPV artifacts.
The SDK is the installable product for day-to-day work across ODPS, ODPC, ODPG, ODPV, and draft ODPR. It gives teams one command surface for validation, inspection, generation, portfolio workflows, contract checks, and MCP access.
$ pip install open-data-products
The companion specification repositories also include source-level utilities for artifact generation, validation, search, graph reasoning, vocabulary alignment, and draft recipe checks.
Build, validate, explain, search, regenerate, and check catalog artifacts.
Validate, convert, summarize, traverse, analyze, extract agent context, and render graph explorers.
Generate vocabulary artifacts, validate terms, search, resolve agent context, and detect cross-spec drift.
Validate recipes, search recipe patterns, regenerate artifacts, and check agent-facing workflow resources.
Don't just take our word for it. Hear from the many organizations and professionals who have transformed their data strategies with Open Data Products, starting from the Open Data Product Specification (ODPS) foundation.
Here's what some of our satisfied users have to say about their experience:
"ODPS makes sense if you have multiple data products requiring consistent governance, want to scale governance without headcount, work in data mesh architectures, or want to provide structured metadata to your AI agents"
Michael Eichenseer
Data Engineer, Codecentric
"I'm building a UX PoC, v4.1 is a perfect balance between business needs & goals, formalizing a data interfaces/data layer and create a interoperable dataproducts."
Eugene Istomin
Founder, Memex.Team
"ODPS makes product thinking concrete for data and AI — explicit metadata, clear interfaces, lifecycle discipline, and governance by design."
Ron Tolido
CTO and Executive VP, Techstraordinary
"We at BASF built our group-wide Data Product concept largely on ODPS and extended it to our specific needs. Reasons to choose ODPS were its maturity especially in complex non-technical aspects."
Manfred Sorg
Group Data Office, BASF
"Infrastructure, security, persistency, services and processes all have their 'as-code' open standards within the Ferris platform. Now this stack is complete in adding ODPS as a unified standard for regionally distributed data owners."
Thomas Debus
Founder & CTO, Ferris AI
"ODPS fosters collaboration. With standardized data descriptions, we can easily integrate with data marketplaces, reaching a wider audience and accelerating innovation"
Rajanikanth Thouta
Executive Director, iSpatial Techno
Solutions
"Extending X-Road OAS service descriptions with ODPS makes data product-related metadata available in the service catalogue more unified, improves its quality, and enables more automation in the maintenance process."
Petteri Kivimäki
CTO, Nordic Institute for Interoperability
Solutions
"ODPS helps our platform to bridge the gap of discoverability and use of data products in the context of data spaces."
Lukas Hohl
Co-Founder & CPO, Beyond Civic
"ODPS provides a solid foundation for building robust and efficient generative Al workflows and pipelines"
Elias Helou
Founder, VesoAI
We'd love to hear your story! Share your insights and let us know how the standards family has made a difference for
you.
Your experiences can inspire others and help us enhance our offerings.
Find the specifications, schemas, examples, repositories, and SDK resources for the Open Data Products standards family.
ODPS remains the foundation, with ODPC, ODPG, ODPV, draft ODPR, and the Python SDK extending the family into catalogs, graph relationships, shared vocabulary, repeatable workflow contracts, and AI agent workflows.
Your playbook for defining, cataloging, connecting, governing, and launching data products, now made practical. Dive into real-world guidance, clear documentation, and step-by-step examples that help you move beyond theory and into execution.
Explore standards family examples to see how data products can be structured, cataloged, connected, and interpreted for human and AI workflows.
These examples separate public adoption signals from plausible operating scenarios. The first group is grounded in public references. The second group shows practical patterns organizations can implement with ODPS, ODPC, ODPG, ODPV, and draft ODPR.
ODPS is included in NATO's Data Centric Reference Architecture for the Alliance, where it is referenced as a specification supporting Data-as-a-Product principles in federated data environments.
Read NATO reference articleAlation positions its Data Products Builder and Marketplace around governed, machine-readable data products based on ODPS, with support for business users and AI agents.
View Alation BuilderMaysano describes itself as an operating layer for data products, governance, and AI, using structured standards to connect business outcomes, product accountability, reusable governance profiles, and execution.
Read Maysano publicationFIWARE and ODPS announced collaboration to integrate reusable, interoperable data products with FIWARE's Business API Ecosystem for publishing, lifecycle management, and monetization.
Read FIWARE announcementNIIS describes how ODPS can extend X-Road OAS service descriptions so data product metadata becomes more unified, higher quality, and easier to automate in service catalogs.
Read NIIS articleBeyond Civic lists ODPS support in its product release history, including ODPS 3.1, ODPS 4.0, and planned ODPS 4.1 support for AI-ready data collaboration.
View Beyond Civic productBASF's public testimonial says its group-wide data product concept was built largely on ODPS and extended to fit enterprise needs, especially complex non-technical aspects.
Read ODPS adoption articleA data office uses ODPS for product definitions, ODPC for portfolio grouping, ODPG for value and dependency mapping, and ODPV for consistent terms across domains.
A platform team asks providers to publish ODPS files before listing products, making licensing, pricing, SLA, data quality, access, and ownership visible before consumption.
An agent searches product specs, catalogs, graph links, and vocabulary sidecars before deciding whether a product is useful, trusted, licensed, and fit for a business task.
Reusable SLA, data quality, access, contract, and pricing profiles are referenced across products so governance becomes reviewable metadata instead of scattered policy text.
A portfolio team links products to use cases, objectives, KPIs, signals, systems, and agents so leaders can see which products support which business outcomes.
Draft ODPR workflow contracts declare validation, localization, review, release, and agent-safe automation steps so teams can repeat data product delivery with fewer hidden manual habits.
Observed signals should be used as evidence. Plausible scenarios are implementation patterns that show where the standards family can be applied without implying a named organization has already done that exact workflow.
We want to hear from you! Share your insights with us, and let's collaborate to transform your vision
into reality.
Your feedback could lead to exciting new possibilities!
The Open Data Products standards family is developed for practical adoption by organizations, platforms, consultancies, developers, and AI agent builders.
Use the standards in your data product operating model, implement them in platforms, validate files with the SDK, or contribute feedback through GitHub.
The Open Data Products standards family is developed and maintained by people working across data products, governance, interoperability, AI readiness, and open standards.
Dr. Jarkko Moilanen
Igniter and maintainer, ODPS, ODPC, ODPV, ODPR, and AI Agent SDK
Manfred Sorg of ODPS
Maintainer
DSc. Toni Luhti
Commercial operations
Tekla Wannas
Ecosystem & Marketing
Antti Poikola
Data Architecture Specialist
Zaher Abou Shakra, PMP®
Maintainer, ODPG and graph-based value relationships
Elias Helou, Veso AI
Applying Sovereign, Secure & Useful AI
Subscribe to our blog for the latest updates, best practices, and insights about the standards family.
