Intellectual Property Rights in Real Estate Data and Databases

Real estate data and databases — from Multiple Listing Service (MLS) compilations to automated valuation model (AVM) training sets — sit at the intersection of property law, copyright doctrine, contract enforcement, and emerging federal regulation. Understanding which intellectual property rights attach to data collections, who holds those rights, and under what conditions they transfer or expire governs how brokerages, proptech firms, and individual agents may legally use, license, or republish structured property information. This page examines the doctrinal framework, classification boundaries, and practical tensions that define IP protection for real estate data in the United States.

• Definition and scope
• Core mechanics or structure
• Causal relationships or drivers
• Classification boundaries
• Tradeoffs and tensions
• Common misconceptions
• Checklist or steps
• Reference table or matrix
• References

Definition and scope

Intellectual property rights in real estate data and databases refer to the legally recognized interests that control the reproduction, distribution, transformation, and licensing of structured collections of property-related information. These interests arise under three principal federal regimes: copyright law (17 U.S.C. §§ 101–1332), trade secret law as federally codified by the Defend Trade Secrets Act of 2016 (18 U.S.C. §§ 1831–1839), and — to a lesser degree — patent law for software-implemented processes under 35 U.S.C. §§ 1–390.

The scope of coverage extends beyond the raw data points themselves (address, price, square footage) to encompass the selection, coordination, and arrangement of those data points within a structured system. A database containing 4 million property records compiled under a proprietary taxonomy qualifies for copyright protection in its organizational structure, even when the individual facts embedded in it do not. This distinction — protection for expression and selection rather than facts — is the operative boundary that the U.S. Supreme Court established in Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340 (1991) (full opinion via Cornell LII).

The intellectual property in real estate overview provides foundational context for how these doctrines apply across the broader real estate sector. For a focused treatment of MLS-specific rights, see MLS database intellectual property rights.

Core mechanics or structure

Copyright protection for databases turns on originality in selection and arrangement. Under Feist, 499 U.S. at 345, the Copyright Act requires "at least some minimal degree of creativity" — a threshold that a mere alphabetical telephone directory fails but a curated property database with proprietary classification logic can satisfy. The copyright attaches at the moment of creation without registration, though registration with the U.S. Copyright Office is required before a plaintiff may bring an infringement suit under 17 U.S.C. § 411, and registration within 3 months of publication preserves eligibility for statutory damages up to $150,000 per work under 17 U.S.C. § 504(c).

Trade secret protection operates under a different logic: it applies when data collections derive independent economic value from not being generally known and are subject to reasonable security measures. The Defend Trade Secrets Act (18 U.S.C. § 1839(3)) defines a trade secret to include "all forms and types of... compilations... whether tangible or intangible" — language broad enough to cover a bespoke AVM training dataset or a proprietary comparables database that a brokerage treats as confidential. Unlike copyright, trade secret protection has no fixed term; it persists as long as secrecy is maintained. The real estate trade secrets reference page examines that protection mechanism in greater depth.

Contractual layers — end-user license agreements (EULAs), MLS subscriber rules, and data-sharing API terms — often provide the most operationally significant constraints on data use, because they bind parties regardless of whether copyright or trade secret claims could independently succeed. The National Association of Realtors® (NAR) publishes MLS rules governing data display, feed access, and syndication rights that function as a private governance regime layered on top of federal IP law (NAR MLS Policy, Policy Statement 7.58).

Causal relationships or drivers

Three structural forces drive the complexity of IP rights in real estate data:

Data aggregation economics. Building a comprehensive real estate database requires assembling records from county recorders, tax assessors, MLS systems, and proprietary survey data. The capital investment in aggregation — estimated at hundreds of millions of dollars for national-scale data providers — creates commercial pressure to assert robust IP claims. The U.S. Copyright Office has acknowledged in its Circular 66 (Copyright Registration for Online Works) that database originality is fact-specific and highly contested.

The Feist gap. Because facts are not copyrightable, raw property data (parcel number, recorded sale price, lot size) sourced from public records cannot be monopolized through copyright. This gap motivates data firms to shift protection strategies toward trade secrets, contractual restrictions, and technical access controls under the Digital Millennium Copyright Act's anti-circumvention provisions (17 U.S.C. § 1201).

Platform concentration. The consolidation of real estate data infrastructure into a small number of dominant platforms — notably CoStar Group for commercial data and Zillow/Realtor.com in the residential segment — concentrates IP control and creates licensing dependencies for downstream users. The Federal Trade Commission has examined data concentration in real estate technology markets as part of its broader scrutiny of data broker practices (FTC Report: Data Brokers (2014)).

Classification boundaries

IP protection for real estate data does not operate uniformly. Four distinct categories define the protection landscape:

Category 1 — Raw public record facts. Parcel IDs, recorded deed prices, tax assessments, and zoning designations are facts drawn from government records. No copyright protection attaches under Feist. These data points may flow freely once obtained from public sources.

Category 2 — Curated proprietary compilations. A database that selects, organizes, and enriches raw facts using a proprietary schema — adding standardized field definitions, quality-scored geocoding, or normalized unit measurements — gains thin copyright protection in its expressive architecture. Reproduction of the architecture, not the underlying facts, constitutes infringement.

Category 3 — Derived analytical products. Automated valuations, price indices, market trend scores, and predictive risk ratings are transformed works. Copyright in these products vests in the authoring entity for the creative expression embedded in the analytical methodology, though the underlying facts feeding the model remain unprotected. For AI-generated variants, copyright eligibility is unsettled; the U.S. Copyright Office's Guidance on AI-Generated Works (Federal Register Vol. 88, No. 51, March 2023) clarifies that works lacking human authorship do not receive copyright.

Category 4 — Software and APIs. The database management system, query logic, and API architecture enabling data access are protectable as software under 17 U.S.C. § 101 and potentially patentable for novel, non-obvious technical methods under 35 U.S.C. § 101 subject to Alice Corp. v. CLS Bank International, 573 U.S. 208 (2014) (opinion via Cornell LII). The real estate software patent landscape page addresses patent eligibility issues in depth.

Tradeoffs and tensions

Openness vs. protection. Public interest in transparent real estate markets — enabling fair housing compliance, neighborhood-level price research, and urban planning — conflicts with private IP claims that restrict data access. The Fair Housing Act (42 U.S.C. § 3604) and the Consumer Financial Protection Bureau's data requirements under the Home Mortgage Disclosure Act create regulatory mandates for disclosure that can conflict with database licensors' access restrictions.

Thin copyright's enforcement limits. Thin copyright over database structure provides narrow protection: only virtually identical reproduction infringes. This makes it difficult to stop a competitor who independently re-compiles public records into a structurally similar but textually distinct database — a practice sometimes called "sweat of the brow" replication that U.S. law, unlike European Union database rights under the EU Database Directive 96/9/EC, does not categorically prevent.

Contract vs. copyright. Relying on EULA-based access restrictions rather than copyright can produce outcomes that override the limitations Congress built into copyright law — including fair use under 17 U.S.C. § 107. Courts in the Ninth Circuit have permitted contractual claims to restrict uses that copyright alone would not block, creating an asymmetric enforcement environment. The real estate data intellectual property page analyzes how these conflicts develop in practice.

Common misconceptions

Misconception 1: All publicly available real estate data is free to republish. Public availability means the data is accessible, not that reproduction is unrestricted. County assessor databases are public, but a data aggregator that re-packages them into a proprietary database can assert copyright in its organizational expression and enforce EULAs against downstream users.

Misconception 2: Copyright in a database protects the individual data points within it. Copyright protects only the selection, arrangement, and coordination — not the facts. Extracting raw price or parcel data from a copyrighted compilation and publishing those facts separately does not infringe the copyright, provided the extraction does not reproduce protected expressive architecture.

Misconception 3: MLS data belongs to the listing agent. Under NAR's model MLS rules, agents grant the MLS a license to use listing content, but ownership of the compiled database typically vests in the MLS organization. The agent's copyright in original descriptive text (see real estate listings) is distinct from the MLS's compilation rights.

Misconception 4: Trade secret protection requires registration. No registration process exists for trade secrets under either federal or state law. Protection arises automatically when the information has commercial value from its secrecy and the holder takes reasonable steps — such as employee NDAs, access controls, and compartmentalized internal systems — to maintain that secrecy.

Misconception 5: AI-generated data analyses are automatically copyrighted. Following the Copyright Office's 2023 guidance, works produced without sufficient human authorship — including AI-generated property valuations or market reports produced without creative human input — do not qualify for copyright registration. The real estate AI tools copyright issues page examines this area further.

Checklist or steps

The following sequence identifies the key analytical steps for determining IP rights in a specific real estate data or database asset:

1. Identify the data type — Classify the asset as raw public record facts, curated compilation, derived analytical output, or software/API system, since each category carries different protection eligibility.
2. Assess originality — Determine whether selection, coordination, or arrangement of the data exhibits minimal creativity beyond mechanical aggregation under the Feist standard.
3. Locate the creation chain — Identify who created the database (employee vs. independent contractor) and whether any work-for-hire agreements under 17 U.S.C. § 101 allocate copyright to an employer entity.
4. Review existing licenses — Examine all upstream data licenses, MLS subscriber agreements, and API terms of service for field-of-use restrictions, syndication prohibitions, and sublicensing rights.
5. Confirm registration status — Check the U.S. Copyright Office Public Catalog for registration records, noting that registration within 3 months of publication affects available remedies.
6. Evaluate trade secret eligibility — Document security measures in place: access logs, NDAs, encryption protocols, and internal distribution controls.
7. Map patent landscape — Search the USPTO Patent Full-Text Database for issued patents covering the data processing methods embedded in the system.
8. Identify contractual gaps — Determine whether copyright or trade secret protection would survive termination of the governing contract, since some protections survive while others depend entirely on the contractual relationship.
9. Document chain of title — Confirm written assignments for any IP created by contractors or third-party vendors under the framework detailed in real estate IP assignment agreements.
10. Assess fair use or public interest exceptions — Evaluate whether uses such as research, commentary, or compliance reporting qualify under 17 U.S.C. § 107.

Reference table or matrix

References

• U.S. Copyright Office — Title 17, United States Code
• U.S. Copyright Office — Circular 66: Copyright Registration for Online Works
• [U.S. Copyright Office — AI-Generated Works Guidance, Federal Register Vol. 88 No. 51 (March 2023)](https://www.federalregister.gov/documents/2023/03/16/2023-05321/