A data space is a trustworthy, decentralised and federated ecosystem where organisations, and individuals, share private or protected data and services—voluntarily, and with strict and sovereign control over access and reuse—to create new economic, social, and societal value.
What it includes
A Community of Practice
Gathers all types of organisations such as
SMEs, large firms, public bodies, NGOs, and research actors. A Data
Space Governance Authority (DSGA) may represent the community.
A governance framework
A shared Rulebook capturing principles, standards, policies,
agreements, enforcement, and conflict-resolution for both business and
technology in a human and machine-readable format.
A data-sharing infrastructure
Modular building blocks, operated
decentrally by participants or centrally at data space level:
collaboration tools, identity & trust, catalogues for discovery,
vocabulary hubs, marketplaces, contracting, connectors for exchange,
policy engines, monitoring, billing, and more.
Key principles
Control & sovereignty
Data holders decide who uses what, for which
purpose.
Privacy & security by design
Aligned with GDPR and related EU rules
Interoperability
Common models, standards, and protocols.
Fair value-sharing
Reuse can be free or compensated, with transparent
terms.
Benefits
Faster collaboration
Shared tools and policies reduce negotiation
time.
Better quality & compliance
Conformance checks and audits build
trust.
Scalability & efficiency
Reusable blocks replace one-off
integrations.
AI-ready foundation
Safe pooling of private/commercial data for
trustworthy, agentic AI across organisations and sectors.
Sovereignty
Control over access and reuse at participant and
domain/sector level.
Network effect
Sharing is first enabled within a domain (intra-data
space) then across multiple domains (inter or cross-data space) to
form a European “network of networks”
What a data space is not
Not open data
Access and reuse is controlled in a data space, and it
follows explicit rules; sharing may be remunerated.
Not data lake
Data can stay with the holder; exchange is federated
between participants and not centralised in one place.
Not just policy talk
It runs live services (identity, catalogues,
connectors) that enable concrete use cases.
The Data Space Cookbook
Who starts a data space?
A community of practice, often a mix of public and private actors, decides to collaborate. It can be sectoral or cross-sectoral, EU-level, national, or local.
1) Define scope & mission
- Write a clear Mission Statement and success criteria.
- Frame initial use case boundaries and data types (personal & non-personal).
Output: One-page mission, early use case list.
2) Set up governance
- Constitute a Data Space Governance Authority (DSGA)—optionally/preferably backed by a legal entity (association, cooperative, etc.).
- Define decision rights, roles, and onboarding/offboarding.
- Draft the Rulebook (principles, policies, standards, agreements, enforcement, dispute resolution).
- Map applicable laws & soft law (GDPR, DGA, Data Act, sectoral laws, codes of conduct, standards, etc.).
- Specify access rights and data-management practices, including inter-data space sharing rules.
Output: Rulebook v1, participation terms.
3) Map the landscape
- Engage EU/national authorities and sector coordinators.
- Connect with peer data space initiatives (same and adjacent domains and sectors).
- Align with standards/support bodies.
Output: Stakeholder map & alignment plan.
4) Co-design use cases
- Adopt a practical co-design method for intra- and inter-data space scenarios.
- Keep the ecosystem open & inclusive (public/private orgs and end-user communities).
- Formalise agreements on scope, data, roles, and KPIs.
Output: Use case canvases with owners, milestones, and metrics.
5) Build a sustainable business model
- Define value exchange: how data/services are compensated.
- Fund infrastructure and governance operations. Infrastructure operations can require the creation of a dedicated Operating Company of the data space.
- Set incentives for broad participation (incl. SMEs).
Output: Financial model, pricing/compensation options, budget.
6) Build the technical stack
- Choose cross-sector standards (identity/trust, catalague, contracts/consent, policies/rules).
- Select sector ontologies and vocabularies and define pilot ontologies of the data space.
- Favor open standards/protocols for interoperability.
- Implement core building blocks: connectors (as-a-Service or self hosted), identity & trust framework, federated catalogue, vocabulary hub, contracting, policy engine, monitoring, billing, quality.
- Provide a collaboration UI/platform/tools (marketplace, use-case factory, contract negotiation, social network and virality), run by multiple data space participants or by a single Operating Company mandated by the DSGA.
- Add middlewares for specific needs (digital twins, agentic AI frameworks, non data sources such as Open Data and classical marketplaces).
Output: Reference architecture, MVP infrastructure, runbooks.
7) Fund the journey
- Pool resources from founders.
- Secure public/private funding for infra and use cases through EU/national/local call for tenders and proposals, or from VC/corporate funding.
Output: Funding plan & commitments.
8) Operate & grow
- Run the data-sharing infrastructure and community processes.
- Support self-hosted connectors or provide Connectors-as-a-Service.
- Maintain Rulebook, semantic hubs and catalogues.
- Recruit new participants; scale intra and inter data space use cases.
- Track SLIs/SLOs, security & compliance; report on impact and iterate.
Output: Operations dashboard, conformance reports, roadmap.