Common Ontology for Real Assets
Real assets data has always required reconciliation. What has changed is who reads it.
CORA is a vendor-neutral ontology for real estate and infrastructure data. It defines a shared semantic layer across asset classes, ownership structures, capital stacks, geographies, and time, so existing standards can be interpreted together rather than reconciled manually.
Being formed as an open, multi-stakeholder industry collaboration.Global by design.Publicly available.Openly licensed.
What CORA Is
The real assets industry does not lack data standards. It lacks a common semantic layer that allows those standards to be interpreted together.
Over the past two decades, multiple credible standards have emerged for specific audiences, workflows, and reporting obligations. CORA is being formed to connect them: not as another exchange format, but as an ontology that defines shared classes, relationships, identifiers, and provenance primitives for real estate and infrastructure data - across equity, debt, infrastructure (including data centers, renewables, and transport), and operating contexts.
CORA sits above existing exchange standards as a semantic layer. Its purpose is not to replace them, but to make their meanings interoperable.
Horizontal semantic standards move models across analytics, BI, and AI tools. CORA defines the real assets meaning those tools need to preserve.
CORA Is Designed As
- A common ontology for real estate and infrastructure data
- A semantic layer above existing data exchange standards
- A shared model for entities, relationships, identifiers, and provenance
- A foundation for agent-readable real assets data
- Global by design, with extension points for local market and regulatory variation
- Openly governed, openly licensed, and publicly available
CORA Is Not
- A new transport protocol
- A replacement for existing standards
- A vendor product
- A compliance mandate
- A closed membership initiative
- A standard limited to one asset class, geography, or layer of the capital stack
Example fragment
An asset represented in CORA carries cross-standard identifiers, ownership and management relationships, jurisdictional context, and provenance back to source systems. The same record reads consistently whether it is referenced by an operator, an asset manager, a fund administrator, or an allocator.
{
"@context": "https://coradata.org/v0.1/context.jsonld",
"@type": "Asset",
"assetClass": "Multifamily",
"identifiers": {
"oscre": "OSC-9182-A",
"mits": "MITS-77123",
"redi": "REDI-PROP-2024-014"
},
"ownedBy": { "@type": "InvestmentVehicle", "@id": "fund:opp-fund-iii" },
"managedBy": { "@type": "AssetManager" },
"operatedBy": { "@type": "Operator" },
"locatedIn": { "@type": "Jurisdiction", "@id": "jurisdiction:us-tx-dallas" },
"provenance": {
"sourceSystem": "PropertyManagementSystem",
"asOf": "2026-03-31"
}
}Illustrative. A common semantic layer allows the same real asset to be interpreted across operations, ownership, investment reporting, debt, and governance contexts. The v0.1 specification will publish the full set of classes, relationships, identifiers, and provenance primitives.
Scope of v0.1
Real assets data does not break in one place. It breaks across three recurring dimensions: the capital stack, the asset class, and the jurisdiction.
CORA v0.1 is designed to model those intersections. Data aggregates up from assets to portfolios, funds, and allocators. It disaggregates back down for valuation, operations, risk, reporting, and audit. At every handoff, meaning changes unless it is explicitly modeled.
The problem compounds because the planes intersect.
A global investor does not manage geography, asset class, and capital structure as separate planes. A vertically integrated real estate firm does not either. The same semantic friction appears inside the organization, across regions, business lines, acquired entities, and operating systems.
CORA v0.1 begins where the industry needs common ground: a global foundation, stable enough to coordinate across contexts and extensible enough to support sector and jurisdictional variation over time.
Real estate, equity
Multifamily, office, industrial and logistics, retail, hospitality, healthcare and senior housing, student housing, self-storage, and mixed-use. Direct property and portfolio structures, fund-level entities, ownership hierarchies, lease and operational data, valuation, transactions.
Real estate, debt
Mortgages, CMBS, mezzanine and preferred equity, construction lending, and other loan-level instruments. Servicer and lender entities. Securitization structures.
Infrastructure
Data centers, renewable energy assets, transport infrastructure, with defined extension points for additional sectors.
CORA spans the full real assets capital stack: capital allocators, fund sponsors, asset managers, owner-operators, property managers, fund administrators, lenders, servicers, and technology platforms. The ontology models the entities, relationships, and data flows required to consolidate information from property-level operations up to portfolio, fund, and investor reporting, and to disaggregate it back down for analysis, operations, risk, and audit.
CORA is global by design. Real assets data changes meaning across jurisdictions: legal entity structures, lease conventions, valuation standards, regulatory regimes, currencies, and reporting obligations all vary by market. CORA provides extension points for regional and jurisdiction-specific requirements across North America, EMEA, APAC, and Latin America. Specific mappings to regulatory and industry frameworks, including AIFMD, SEC, INREV, NCREIF, and regional equivalents, will be developed through the Standards Mapping working group.
Subsequent versions extend coverage through community contribution and open committee approval.
Currently in Formation
The formation period is focused on four workstreams: confirming founding-period participants, drafting the v0.1 specification, establishing open governance, and mapping CORA to existing real assets standards.
Why Now
Reports tolerated semantic drift. Agents do not. They read data directly and reason from it, without the institutional judgment that human readers apply by default - which is why the next generation of real assets infrastructure cannot rely on human interpretation as the integration layer.
Get Involved
CORA is being formed in the open with institutional, technical, and standards-community participation. Four paths exist during the formation period:
Submissions are recorded privately. A member of the formation group will respond directly.