Algorithmic Rent Pricing in 2026: Revenue Management Software You Can Audit

A rent recommendation can look reasonable on a dashboard and still create a serious compliance problem. Here is how to build pricing systems with clear data boundaries, explainable recommendations and a complete audit trail.

4 state laws compared DOJ settlement terms 5-layer compliant architecture CIO audit checklist
Key takeaways
Algorithmic rent pricing is not banned nationwide. Laws target coordination and competitors' nonpublic data.
New York, California, Connecticut and New Jersey now regulate it, each in a different way.
The DOJ's proposed RealPage settlement reads like an engineering spec for compliant pricing.
Compliance is architectural: data classification, tenant isolation, explainable AI pricing and audit trails.

The risk often has little to do with the mathematical model itself. It comes from where the pricing data originated, whether competitors' confidential information entered the system, or whether the software is coordinating decisions across independent property owners.

That distinction now sits at the centre of algorithmic rent pricing in the United States. Through 2025 and 2026, New York, California, Connecticut and New Jersey introduced or strengthened laws that target algorithmic rent setting and the use of competitors' sensitive data. Cities have passed their own bans, federal enforcement has reached both pricing software providers and major property managers, and a federal court has already paused part of one state law.

For property management companies and PropTech platforms, the challenge is now architectural. Revenue management software needs to support independent pricing decisions with clear data boundaries, traceable inputs and recommendations that a revenue manager can explain after the fact.

The goal is not to remove algorithms from rent pricing. It is to make sure the data and logic behind every recommendation can withstand scrutiny, and that people stay in control of the final decision.
01 · REGULATION

Algorithmic Rent Pricing Regulations: What Changed in 2025 and 2026?

The regulatory landscape is not a single nationwide ban on rent pricing algorithms. Each state has taken a different approach, and their requirements should not be treated as interchangeable.

21
states with bills targeting algorithmic rent setting (July 2026)
4
states with enacted algorithmic rent pricing laws
$7M
LivCor settlement with nine attorneys general
12 mo
minimum age of nonpublic training data under the proposed RealPage settlement
NY New York
Partly enjoined
Section 340-b targets software that coordinates competing landlords. The landlord-facing provision is paused by a 2026 injunction.
CA California
In force
AB 325 covers common pricing algorithms in every industry, using public or nonpublic competitor data.
CT Connecticut
In force
Section 47a-4f bans revenue management devices that use nonpublic competitor data, even anonymised.
NJ New Jersey
Jul 1, 2027
The FAIR Act restricts "coordinators" and bans parallel pricing coordination.

New York: Section 340-b and the 2026 Preliminary Injunction

New York signed Senate Bill S7882 into law on October 16, 2025, adding Section 340-b to its Donnelly Act antitrust statute. The law took effect on December 15, 2025. It prohibits knowingly or recklessly facilitating agreements between residential landlords not to compete, including through software and algorithmic devices that perform a coordinating function between competing landlords.

The legal status has since changed. In September 2026, a federal judge granted RealPage a preliminary injunction blocking enforcement of Section 340-b(3), the provision that treats a landlord's use of covered software recommendations as an unlawful agreement. The New York Attorney General has asked the court to reconsider.

Important
The injunction is not a safe harbor. The prohibition on facilitating actual anticompetitive agreements remains enforceable, and the case is still ongoing. Teams operating in New York should treat the law as active risk rather than settled relief.

California: AB 325 and Common Pricing Algorithms

California's AB 325 was signed on October 6, 2025, and AB 325 took effect on January 1, 2026. It amends the Cartwright Act to prohibit using or distributing a "common pricing algorithm" as part of an agreement that restrains trade, and to prohibit coercing another party to adopt a price or commercial term the algorithm recommends.

Two details matter for technology teams:

1.It is not limited to housing. AB 325 applies to pricing algorithms in every industry operating in California, not only residential rent.
2.It does not separate public and nonpublic data. A common pricing algorithm is any methodology used by two or more parties that uses competitor data to recommend or influence a price. The definition covers both public and confidential competitor data.

A shared algorithm does not automatically create a violation. A conspiracy or coercion is still required. But how a pricing product is distributed, configured and used across competing customers now carries direct legal weight.

Connecticut: Revenue Management Device Restrictions

Connecticut's Public Act 25-1 introduced Section 47a-4f, effective January 1, 2026. It prohibits using revenue management devices that calculate nonpublic competitor data about rents or occupancy to advise landlords on rental rates or whether to leave units vacant.

Connecticut was the first state to limit its prohibition to nonpublic competitor information, which leaves room for public data in rent pricing. Its definition still covers competitor information even when it has been anonymised. The law also separates revenue management recommendations from aggregated reports that publish existing rental data without recommending future prices or occupancy levels.

New Jersey: The FAIR Act

New Jersey signed the Forbidding the Algorithmic Inflation of Rent (FAIR Act) on July 20, 2026, making it the fourth state to expressly regulate algorithmic rent pricing. The law takes effect on July 1, 2027.

The FAIR Act restricts "coordinators" that operate algorithmic revenue management software and prohibits parallel pricing coordination, which closes the gap for manual workarounds that replicate what an algorithm does automatically. For property technology providers, the effective date gives roughly a year to review data pipelines, data sharing arrangements and pricing workflows.

How the four state laws compare

State
Law
What it targets
Public data
Status
New York
GBL § 340-b (S7882)
Software performing a coordinating function between competing landlords
Not distinguished
Effective Dec 15, 2025; § 340-b(3) enjoined Sep 2026
California
AB 325 (Cartwright Act)
Common pricing algorithms used in agreements or with coercion, all industries
Covered
Effective Jan 1, 2026
Connecticut
§ 47a-4f (PA 25-1)
Revenue management devices using nonpublic competitor data
Allowed
Effective Jan 1, 2026
New Jersey
FAIR Act
Coordinators and parallel pricing coordination
Not distinguished
Effective Jul 1, 2027

City-Level Bans on Rent-Setting Algorithms

Local rules add another layer. Several give tenants a private right of action with per-violation penalties, and tenant lawsuits under these ordinances began appearing in 2026.

San Francisco San Diego Berkeley Minneapolis Providence Philadelphia Seattle Jersey City

A platform that serves multiple markets therefore needs pricing controls that can be configured by jurisdiction, not one global setting.

Rent pricing algorithms must not become infrastructure for competitors to coordinate their market behaviour.
The common concern behind every state and city rule
02 · ENFORCEMENT

The RealPage DOJ Settlement and LivCor: Why Data Provenance Matters

The enforcement actions involving RealPage and major property managers show why regulators look at the data behind revenue management recommendations, not just the recommendations themselves.

January 2025
DOJ files an amended complaint adding six large landlords to its RealPage case.
October 2025
California signs AB 325 (Oct 6) and New York signs S7882 (Oct 16).
November 2025
DOJ announces a proposed settlement with RealPage.
December 15, 2025
New York's Section 340-b takes effect.
January 1, 2026
California AB 325 and Connecticut Section 47a-4f take effect.
June 18, 2026
Nine attorneys general announce a $7 million settlement with LivCor.
July 20, 2026
New Jersey signs the FAIR Act.
September 2026
Federal court blocks enforcement of New York's Section 340-b(3).
July 1, 2027
New Jersey's FAIR Act takes effect.

In January 2025, the US Department of Justice filed an amended complaint against six large landlords, expanding its antitrust case against RealPage. The allegations concerned the use of competitors' competitively sensitive information through common pricing algorithms to reduce competition in rental markets.

In November 2025, the DOJ announced a proposed settlement with RealPage. The proposed terms read like an engineering specification for algorithmic pricing compliance:

01
No live competitor data. The software may not use competitors' nonpublic, competitively sensitive information when generating runtime pricing recommendations.
02
Aged training data only. Model training is limited to historical nonpublic data that is at least 12 months old.
03
No aligned recommendations. The software may not make identical pricing recommendations to different owners in the same market, and features that limit price decreases must change.
04
Human control by default. Auto-accept pricing features must be removed unless a user manually opts in.
05
Independent oversight. A court-appointed monitor reviews compliance.

Enforcement continued into 2026. On June 18, 2026, California and a coalition of eight other states announced a $7 million settlement with LivCor. The settlement addressed allegations that LivCor used RealPage's revenue management system to align rental prices through confidential pricing information. LivCor agreed to stop using software that relies on competitors' competitively sensitive information to set rents.

The lesson for the industry is clear. A property manager cannot evaluate rental pricing software on recommendation accuracy alone. It must also understand the source of the data, the relationships between data contributors and the way the system generates recommendations. That is data provenance, and it is now a compliance requirement.

Free assessment
Not sure what data feeds your pricing engine?
Get a free review of data isolation, pricing traceability and PMS integration readiness.
Book the Free Assessment →
03 · DATA

What Data Should Rent Pricing Software Use?

Property managers still need informed pricing decisions. Occupancy, lease expirations, operating costs and local market conditions all affect revenue.

The important distinction is between information generated by the organisation's own operations and confidential information obtained from competing property owners.

✓ Use
Internal operational data
Portfolio occupancy: current occupancy and vacancy
Lease history: past rates and renewal outcomes
Operating costs: expenses and maintenance
Unit attributes: amenities and availability
Internal demand: lease expiry patterns and trends
! Use with care
Public market comparables
Advertised rents
Publicly listed unit availability
Check collection methods, contracts and terms of service
Shared algorithms using public competitor data can matter under AB 325
✕ Exclude
Competitors' nonpublic data
Confidential rent rolls and occupancy
Lease-level data from other owners
Anonymised competitor data
Data from unapproved sharing arrangements

Independently operated rental revenue management software can draw on internal data such as current occupancy, historical lease rates, property-level operating costs, unit characteristics and internal demand trends.

Publicly available market comparables, such as advertised rents and listed unit availability, can add market context. But public visibility alone does not settle every data-use question. Collection methods, contractual restrictions and terms of service still apply, and California's AB 325 shows that even public competitor data can matter when it feeds a shared algorithm used by competing parties.

Strong pricing data governance therefore separates three categories inside the platform: data generated internally, data that is genuinely public, and data obtained under third-party agreements. That separation should be visible in the system, not left to assumptions about where a number came from. It starts with a clear data strategy before any model is trained.

04 · ARCHITECTURE

How to Build Compliant Revenue Management Software: 5 Architecture Layers

Compliant real estate revenue management software must make it possible to identify every input behind a recommendation. That requires more than a database of rent values. It requires an architecture that preserves ownership, origin, permitted use and transformation history.

1. Source-Level Data Classification

Every incoming dataset should carry metadata identifying its source, collection date, owner, permitted use and geographic coverage.

An internal lease record should be tied to the relevant property portfolio. A public market listing should keep its source and collection timestamp. A third-party dataset should include its usage permissions and, where relevant, the jurisdictions where it may be used.

Classification should happen when data enters the platform. Reconstructing its origin after it has passed through several processing stages is unreliable.

2. Multi-Tenant Data Isolation

Internal operational data must stay within clearly defined organisational boundaries.

A multi-tenant revenue management platform must prevent one property manager's confidential information from entering another customer's pricing workflow. Multi-tenant data isolation should apply to databases, storage, processing jobs, APIs, shared feature stores and model training pipelines. Our guide to multi-tenant SaaS architecture covers the isolation patterns in more depth.

In practice, this means strict access controls, tenant-aware queries, encryption and regular tests for cross-tenant data exposure.

Watch out
Anonymisation is not permission to share competitor data. Connecticut's definition of nonpublic competitor data explicitly covers anonymised information.

3. Controlled Pricing Inputs

The pricing engine should receive only approved datasets.

For each recommendation, the system should identify which internal metrics and permitted public comparables influenced the result. Data from unapproved competitor-sharing arrangements should be filtered out before it reaches the model, and training datasets should follow the same rules as runtime inputs.

A feature-level data inventory helps here. It records which model features depend on which sources, so teams can spot risky dependencies before a model is deployed. Jurisdiction-level rules can then switch specific inputs on or off by market.

4. Explainable AI Pricing

A pricing engine should produce more than a suggested rent.

Explainable AI pricing records the factors behind each recommendation: the property data, market comparables, model version and business rules that applied. For example, a recommendation to raise a unit's asking rent might be supported by a change in internal occupancy, recent lease outcomes on similar units and publicly advertised rents nearby.

Avoid
Black-box recommendation
A single suggested rent with no visible inputs. Reviewers cannot tell what drove it, and the organisation cannot defend it later.
VS
Recommended
Explainable recommendation
A suggested rent with its supporting factors, data sources, model version and rules, ready for a revenue manager to accept, adjust or reject.

This kind of AI pricing transparency gives revenue managers what they need to review, challenge or override a recommendation with confidence. The explanation must be grounded in the actual inputs the system used, not a generated narrative that simply sounds plausible. Our overview of explainable AI explains why this matters in regulated industries.

5. AI Audit Trail for Every Recommendation

Each pricing recommendation should generate an audit record containing:

Pricing audit record Per recommendation
Identifiers
Property and unit identifiers
Timing
Timestamp of the recommendation
Sources
Source datasets and their versions
Model
Model version and configuration
Features
Input features and relevant transformations
Output
Recommended price and applicable constraints
Human decisions
Overrides and approval details
Final result
Final price published to the property management system

This AI audit trail makes it possible to reconstruct how a recommendation was generated and whether the final decision followed the organisation's approved process. It is the foundation of any algorithmic pricing audit, and it works best when it is part of continuous compliance infrastructure rather than a report assembled after a regulator asks.

05 · INTEGRATION

Property Management System Integration: Where Compliance Becomes Operational

Revenue management software rarely operates in isolation. It connects with property management systems (PMS), leasing platforms, CRM applications and reporting tools, and each connection is another place where controls can break.

A recommendation may be generated in one platform, reviewed by a revenue manager and then pushed into the PMS. If the integration does not carry the recommendation's sources and approval details with it, the organisation loses the evidence it needs to explain the final rent.

A stronger property management system integration workflow follows a controlled sequence:

01
Internal property data
From the organisation's own systems
02
Permitted public market data
From approved sources only
03
Validation and classification
Source, owner and permitted use tagged
04
Approved pricing engine
Runs on approved inputs only
05
Explainable recommendation
Price plus supporting factors
Human gate
Human review and business rules
Accept, adjust or reject
07
Authenticated PMS API
Receives only the approved price
08
Published rent and audit record
Stored together
Read left to right, row by row. Nothing reaches the PMS without passing the human gate.

API integrations should use authenticated service accounts, scoped permissions and validation rules. A pricing service should only update the properties and fields it is authorised to manage.

Avoid
Suggested price
A model output. It informs the revenue manager but never changes a live rental rate on its own.
VS
Recommended
Approved price
A decision made by an authorised person under the organisation's rules, sent to the PMS with its audit record.

A model output should not become a live rental rate automatically unless the organisation's policy explicitly allows that workflow. The DOJ settlement terms on auto-accept features point in the same direction.

For a deeper look at connecting these systems cleanly, see our guide to property management system integration and how to avoid API integration debt.

06 · CHECKLIST

What CIOs Should Audit Before Deploying Revenue Management Software

Before selecting or expanding rental pricing software, IT leaders should review the complete data lifecycle. Start with five questions:

Where does the pricing data originate? Identify internal records, public comparables, licensed datasets and any information obtained from external property managers.
Can one customer's data influence another customer's recommendations? Review tenant isolation, model training pipelines, shared feature stores and third-party processing arrangements.
Can every recommendation be reconstructed? Confirm that the platform records source data, transformations, model versions and human overrides.
Who controls the final price? Establish whether the system recommends, approves or automatically publishes changes, and define permissions for each role.
Can the organisation demonstrate independent decision-making? Make sure pricing workflows do not rely on competitor coordination, shared confidential data or features that constrain independent pricing.

These checks belong in vendor evaluation and in ongoing governance. Contractual assurances are not enough if the technical architecture cannot show how the platform handles data. Pricing algorithm compliance has to be provable in the system itself.

07 · SEAFLUX

How Seaflux Helps Build Compliant PropTech Revenue Management Platforms

Seaflux is a PropTech software development partner and AWS Select Consulting Partner that helps property technology companies and property managers build pricing systems that are accurate, explainable and auditable from the start.

Our teams bring together the disciplines that compliant rent optimization software depends on:

Custom real estate software development
We design and build custom real estate software development projects, including property management software development, revenue management software development and tenant platforms built around your workflows.
Data engineering for pricing data governance
Our data engineering services build classified, tenant-isolated data pipelines with source metadata, lineage and jurisdiction-aware controls.
Explainable AI and machine learning
Our AI and machine learning development services create pricing models that record their inputs and explain each recommendation to the revenue managers who make the final call.
MLOps and model governance
Our MLOps services version models, track training data age and keep audit trails consistent across every release.
PMS and API integration
Through custom software development, we connect pricing engines to PMS, leasing, CRM and accounting platforms with scoped, authenticated APIs that carry approval records with every price change.
19%Faster leasing
We have delivered real estate platforms in production, including an AI-powered rental management platform for a Canadian rental SaaS company that reduced leasing time by 19%.
08 · NEXT STEPS

Build Revenue Management That Stands Up to Scrutiny

Algorithmic rent pricing has entered a more demanding regulatory environment, and algorithmic rent pricing laws are still moving. Property managers need revenue management software that supports independent decisions and produces evidence for the data and logic behind every price.

The practical response is an architecture built on internal operational data, permitted public comparables, strict data isolation and traceable recommendations, with people approving the final rent. PMS integrations must preserve those controls when prices move into live leasing workflows.

For real estate technology leaders, the next step is to map the pricing data lifecycle. Identify every source, processing stage, model input, integration and approval point. Any part of the workflow that cannot be explained or audited deserves a closer look.

Free PropTech Data Architecture Assessment
Can your current revenue management platform show exactly which data sources influenced a rent recommendation, and prove that competitor-confidential information was excluded?
A focused architecture review can find the gaps before they become compliance problems. Seaflux offers a Free PropTech Data Architecture Assessment to help property technology teams evaluate their pricing systems.
✓Data isolation ✓Pricing traceability ✓PMS integration readiness

Frequently Asked Questions (FAQ): Get the Answers You Need

Hardik Dangodara

Hardik Dangodara

Business Development Manager

Claim Your No-Cost Consultation!

Let's Connect