The Open Data Product Specification (ODPS) is a vendor-neutral, open-source, machine-readable metadata model for data products.
It defines the key objects, attributes, and relationships that describe digital data products — enabling a consistent, interoperable way to manage, share, and monetize them.
Built on established standards such as schema.org, ODPS applies modern product thinking to data.
It is designed for both open and commercial data products, supporting interoperability, monetization, and AI-native automation across ecosystems.
Version 4.1 introduces the productStrategy object — a new addition that connects data products directly to measurable business goals.
Data products can now declare:
- Objectives — the business intent or OKR objective the product supports.
- Product KPIs — how success is measured at the product level.
- Business KPI (
contributesToKPI) — the higher-level outcome the product drives. - Strategic Alignment — optional references to corporate strategies or initiatives.
This makes ODPS 4.1 the first open standard to formalize the link between data products and business performance.
It supports both KPI and OKR models, turning every YAML file into a living artifact of strategy execution — showing not just what a data product delivers, but why it matters.
Example excerpt
productStrategy:
status: Planned
startDate: 2026-01-01
endDate: 2026-12-31
objectives:
- en: "Enable smarter marketing campaigns"
contributesToKPI:
id: "KPI-OUT-001"
name: "Retail Revenue Uplift"
unit: "percent"
target: 5
direction: increase
timeframe: "2025-Q4"
productKPIs:
- id: "KPI-OUT-001-A"
name: "Customer dataset usage"
unit: "downloads"
target: 4000- Enable interoperability between organizations, data platforms, marketplaces, and tools.
- Reduce metadata conversion friction and errors between systems.
- Accelerate the design, testing, and deployment of data products.
- Power tool automation and DataOps through a shared schema.
- Bridge the gap between technical outputs and business outcomes.
Note: In “Open Data Product,” the word open refers to the openness of the standard, not to open-data licensing or access.
ODPS is adopted by organizations and initiatives across public and private sectors — from governments and data marketplaces to enterprises building AI-native data ecosystems.
If your organization uses ODPS, we’d love to add your logo here.
If you’ve got questions about implementation, contributing, or special features, found a bug?
Submit an issue or propose improvements with a pull request to the dev.
The Open Data Product Specification was originally created by
Jarkko Moilanen and Jussi Niilahti.
The project continues to be maintained and expanded under the
Open Data Product Initiative, hosted by the Linux Foundation since 2024.