Canadian enterprise data teams evaluating property data frequently start with the question: "Should we buy an MLS feed, or a data product like BrightCat?" The question is framed as competitive, but the tools actually solve different problems. This comparison walks through where each one is the right fit, where each one falls short, and why most enterprise use cases end up choosing one definitively rather than running both in parallel.
The two data models
An MLS (Multiple Listing Service) is the broker-to-broker cooperation layer for real estate transactions. Its job is to tell participating brokers what is on the market right now. The feed is structured around currency: listings get added, edited, and removed in near real-time as brokerage activity happens. MLS is the system of record for active brokerage.
BrightCat is a weekly-refreshed property intelligence pipeline. Its job is to capture the full listing lifecycle of Canadian properties over time. Every listing, every price change, every relist, every completed sale. The pipeline links those events to the underlying property using stable matchkeys that survive MLS number changes. The pipeline has operated continuously since 2014 and now covers 5.8 million residential properties and 297,000 commercial properties.
MLS optimizes for currency. BrightCat optimizes for continuity. Both are valid design choices; they serve different use cases.
Side-by-side comparison
| Dimension |
MLS Feed |
BrightCat Pipeline |
| Primary purpose |
Broker workflow, active-market brokerage |
Analytics, AI, retention, underwriting, valuation |
| Refresh cadence |
Near real-time (minutes) |
Weekly, continuous since 2014 |
| Historical depth |
Typically current + recent months |
12 years of continuous weekly history |
| Property identifier |
MLS number — changes on relist |
Stable matchkey — persists across relistings |
| Lifecycle states |
Active / Sold / Expired |
NEW, PRICE CHANGED, DROPPED, RELISTED, SOLD |
| Repeat-sale linkage |
Not provided by default |
194,167 verified pairs in Core product |
| Commercial coverage |
Varies by local board |
297,000+ with dual-listing detection |
| Rental coverage |
Varies by local board |
National Canadian coverage in Rentals product |
| Access requirements |
Typically brokerage licensing required |
Enterprise commercial license |
| Delivery |
RESO Web API, RETS |
Snowflake Marketplace, MCP, flat file |
Where MLS is the right tool
MLS access is the right choice when the use case requires one or more of the following:
- Minute-level currency — active brokerage workflows where a new listing needs to surface to the consumer within minutes of hitting the market.
- Brokerage-authorized data provenance — workflows that require MLS-licensed data for regulatory reasons, such as public-facing property search portals or broker CRM integrations.
- Current-state-only analytics — dashboards showing what is on the market now, with no requirement for historical depth.
- Local-board specialization — deep coverage of one specific regional board at maximum granularity.
Where BrightCat is the right tool
The BrightCat pipeline is the right choice when the use case requires one or more of the following:
- Historical depth — AVM training, HPI series construction, cycle analysis, and any analysis that compares current activity against a multi-year baseline.
- Lifecycle state analysis — knowing that a property is a relist rather than a new listing, that it has had three prior price reductions, or that it sat for 180 days before selling.
- Repeat-sale pairs — the gold-standard input for property price indexing. BrightCat Core contains 194,167 verified pairs.
- Pre-mover intelligence — identifying households 4-12 weeks before a move, which requires the persistent property identifier that MLS numbers cannot provide.
- Cross-provincial analysis — a single standardized dataset covering all ten Canadian provinces, not ten separate board feeds to stitch together.
- Commercial RE with dual-listing signals — sale and lease track unification, likely-transacted/likely-sold/likely-leased flags.
- AI-native access — the MCP Connector lets AI agents query the data through natural language; MLS feeds are not designed for this access pattern.
The matchkey problem
The single biggest reason enterprise use cases move from MLS to BrightCat is the MLS number itself. MLS numbers are assigned at the listing level, not the property level. When a property is delisted and relisted, which happens routinely for properties that do not sell on the first attempt, the relist gets a new MLS number. To the MLS system, the relisted property is a new record; to the underlying reality, it is the same property on its second attempt.
For analytics and AI use cases, this matters. A property that originally listed at $950,000, failed to sell, delisted, waited three months, and relisted at $840,000 looks in MLS data like two separate properties: one that expired and one that is new. In BrightCat's pipeline, the matchkey links them, and the dataset exposes the full lifecycle: $950K list → DROPPED → RELISTED at $875K → PRICE CHANGED to $840K → SOLD at $810K. The difference between those two views is the difference between a working retention model and one that consistently misses relisters.
When to run both
A small number of enterprise buyers run both: MLS for an active-brokerage use case (often a public-facing portal or a broker-side workflow) and BrightCat for the analytics, AI, and retention use cases that sit alongside it. When that is the right answer, the two datasets are complementary rather than redundant. The MLS feed powers the customer-facing surface; BrightCat powers the back-office intelligence. The two do not need to be reconciled because they answer different questions.
Frequently asked questions
Is BrightCat an MLS?
No. BrightCat is not an MLS and does not operate or replace an MLS. The MLS is the system of record for active listings at the moment of sale. It is the broker-to-broker cooperation layer. BrightCat is a weekly-refreshed property intelligence pipeline that captures the full listing lifecycle across time and joins these events with stable matchkeys.
Where does BrightCat data come from?
BrightCat operates a pipeline that has been capturing Canadian property listing and sale events on a weekly cadence since 2014. The pipeline builds property-level time series from this weekly capture, linking events to the same underlying property across relistings and across providers. BrightCat is not a wrapper over an MLS feed. It is a twelve-year longitudinal dataset with its own methodology for standardization, deduplication, and matchkey generation.
Why not just use MLS data?
MLS data is the right tool for active brokerage work: the question of what is on the market right now and what the price is right now. For use cases that require longitudinal history (AVM training, HPI series, pre-mover detection, retention timing), MLS is structurally limited because MLS numbers change on relist and there is no stable identifier linking a property's current listing to its prior listings. BrightCat's matchkey methodology closes that gap.
Do I need both BrightCat and MLS data?
For most enterprise use cases, BrightCat's weekly pipeline covers what MLS access is used for. The exceptions are brokerage workflows that require minute-level currency or regulated brokerage use cases that require MLS-authorized data provenance. BrightCat is the better fit for analytics, AI, retention, underwriting, and investor use cases.
What about data freshness?
MLS data is minute-level fresh for the listings currently in MLS systems. BrightCat data is week-level fresh for the full Canadian pipeline including cross-provider standardization. For enterprise analytics and AI use cases, weekly refresh is generally preferable because the downstream pipeline is not built to consume minute-level updates. For brokerage workflow, minute-level MLS freshness is the right choice.
What's the advantage of BrightCat for AI and analytics specifically?
Three things: stable matchkeys (needed for any time-series analysis), lifecycle states (NEW/PRICE CHANGED/DROPPED/RELISTED/SOLD, where knowing which state a property is in changes the analytical conclusion), and cross-provincial standardization (one schema across all ten provinces instead of ten board-specific schemas).
The right Canadian property data source is the one that matches the job. Brokerage workflows need MLS currency. Enterprise analytics, AI, and retention need BrightCat continuity. Most use cases are definitively one or the other.
BrightCat Data · Canadian property intelligence · Since 2014
Related