How to Evaluate Commercial Real Estate Sites Without CoStar
How to Evaluate Commercial Real Estate Sites Without CoStar
CoStar Group has built a near-monopoly on commercial real estate data intelligence. For large brokerages, institutional investors, and corporate tenants, the platform is often treated as a required utility. For independent CRE investors and smaller operators, the pricing — widely reported to run $20,000 to $50,000 per year for enterprise access — makes it inaccessible.
The important question is not "how do I replicate CoStar" but rather "what data do I actually need to evaluate a commercial site, and where does that data exist publicly?" The answer is more favorable than most people expect.
What CoStar Actually Provides
Understanding what CoStar sells helps identify what can and cannot be sourced elsewhere.
CoStar's core data products include:
- Property listings: Active for-sale and for-lease commercial listings across property types
- Sales comps: Historical recorded sales with price, price per square foot, cap rate, and buyer/seller details
- Lease comps: Executed lease transactions including rent, term, tenant, concessions, and effective rent
- Market analytics: Vacancy rates, absorption, asking rents, and supply/demand trends by submarket and property type
- Demographics: Population, income, daytime population, household characteristics within defined radii
- Traffic counts: Average annual daily traffic (AADT) for retail site evaluation
- Tenant profiles: Retailer expansion plans and site selection criteria
CoStar does not produce most of this data — it aggregates it from public records, broker submissions, and field research. The distinction matters when evaluating alternatives.
What You Actually Need for Most Commercial Evaluations
Not every commercial transaction requires the full CoStar stack. For the majority of CRE investment decisions — particularly value-add acquisitions, retail site evaluations, and industrial repositioning — the critical data reduces to:
- Parcel data: Legal description, lot size, building size, zoning classification, assessor data
- Comparable sales: Recent arm's-length transactions for similar property types in the submarket
- Demographics: Population, household income, daytime population within trade area radii
- Traffic counts: AADT for retail and quick-service use cases
- Zoning: Permitted uses, setbacks, FAR, height limits, overlay districts
- Competition: Existing operators in the trade area for retail and restaurant concepts
Each of these has a free or low-cost public source.
Public Data Sources by Category
Parcel Data
County assessors and appraisers publish parcel data for every commercial property in their jurisdiction. This includes:
- Owner of record and mailing address
- Legal description and parcel identification number (APN)
- Site address and location
- Building square footage (if improved)
- Lot size
- Assessed value (land + improvement)
- Year built
- Property use code (which maps to property type)
- Recorded sale price and date of last transfer
This is the same underlying parcel data that CoStar displays for property details. Most counties make it available for free via their GIS portals or ArcGIS Hub. PropIntel sources commercial parcel data directly from county assessors — no intermediary — and makes it searchable by location, property type, and parcel characteristics.
The key limitation: assessor data reflects recorded information at the time of assessment. It may not reflect recent improvements, partial demolitions, or changes in use that occurred between assessment cycles.
Comparable Sales
In most states, recorded sale prices are public record and appear in county deed records. County clerk offices record deeds at transfer, and most include the consideration amount (sale price). Where deed stamps are used, the price can be derived from the stamp amount even when the recited consideration is nominal.
States where sale prices are confidential (non-disclosure states): Alaska, Idaho, Indiana, Kansas, Louisiana, Maine, Mississippi, Missouri, Montana, New Mexico, North Dakota, Texas, Utah, and Wyoming do not require disclosure of sale price in public records. In these states, comparable sales data requires alternative sources — broker-reported comps, county appraisal district records (which differ from sale prices), or negotiated data access.
For non-disclosure states, the practical workaround is using county appraisal district market value estimates as a proxy, combined with broker-sourced comps where accessible.
For disclosure states, the county clerk's deed database is a complete (if sometimes delayed) record of arms-length sales. The challenge is aggregating and interpreting it: deed records are indexed by grantor/grantee name, not by property type or submarket. Building a comp set requires filtering by parcel ID range, geographic bounding box, or property type code — work that a data platform can automate.
Demographics
The US Census Bureau's American Community Survey (ACS) is the authoritative source for demographic data at the block group and census tract level. The ACS publishes five-year estimates annually covering:
- Population and population density
- Household income (median, mean, distribution by bracket)
- Age distribution
- Educational attainment
- Household size
- Owner-occupied vs. renter-occupied housing
Census data is available at no cost via the Census Bureau's data portal (data.census.gov) and via the Census API. The API returns data for geographic identifiers — census tracts, block groups, counties, zip codes — which can be mapped to a trade area by pulling all tracts within a defined radius of the subject site.
Limitation vs. CoStar: CoStar's demographic data includes current-year estimates and three-to-five year projections derived from proprietary models. Census ACS data lags by one to two years and does not include projections. For most acquisition decisions, the lag is acceptable. For site selection by retailers with multi-year horizon requirements, the projection gap is more consequential.
Daytime population: The Census Bureau's Longitudinal Employer-Household Dynamics (LEHD) data, specifically the LODES (LEHD Origin-Destination Employment Statistics) dataset, provides workplace-location employment data that can be used to estimate daytime population at the census block level. This is free and publicly available, though the interface is less polished than CoStar's daytime population heat maps.
Traffic Counts
State Departments of Transportation publish AADT data for most public roads in their networks. Most state DOTs maintain online GIS portals where traffic count data can be downloaded or queried by location. A partial list of open portals:
- Florida: FDOT Traffic Data portal
- Texas: TxDOT Statewide Planning Map
- California: Caltrans AADT data
- Georgia: GDOT Transportation Data Management System
Coverage is strongest on state-maintained roads and interstates. Local road coverage is less complete. For retail site evaluation, AADT data on the primary access road is typically the critical figure, and it is almost always available for arterials and state highways.
Limitation: State DOT AADT data is updated on irregular cycles (often every 2-4 years for lower-volume roads). CoStar licenses real-time traffic volume data from vendors like StreetLight Data. For grocery-anchored retail or QSR site selection where precise traffic is a gating criterion, the timeliness gap matters.
Zoning
Municipal and county GIS portals publish zoning layers as downloadable shapefiles and via ArcGIS Map Services. Coverage has improved dramatically over the past decade — most jurisdictions with populations above 10,000 have a public-facing GIS zoning layer.
The most useful zoning resources beyond the GIS layer:
- Municode: Most municipalities publish their zoning ordinances on Municode (
library.municode.com), searchable by jurisdiction. This is where to find permitted uses by zone, dimensional standards, overlay districts, and special exception processes. - State enabling statutes: State zoning enabling acts define the outer limits of local zoning authority — useful for understanding what variances are available and what processes are required.
- Comprehensive plans: Future land use maps, available from most county planning departments, indicate the intended future zoning designation — relevant for value-add plays that depend on rezoning potential.
Competition Mapping
For retail site evaluation, the composition of the trade area matters. Free and accessible sources:
- Google Maps API: Place search returns business listings within a radius for a given category. The API is free up to a usage threshold; paid above that. Sufficient for preliminary competitive analysis.
- SBA business data: The SBA publishes business location data by NAICS code and geography, useful for understanding the density of competing business categories.
- County business tax receipt / occupational license records: Most Florida counties and many others publish active business license records as open data. These are the authoritative source for active businesses at specific addresses — more reliable than Google Maps for identifying all operators, including those without a strong online presence.
The Aggregation Gap
Every data source described above is genuinely free and publicly available. The practical challenge is not access — it is assembly. Pulling parcel data from a county GIS portal, scraping DOT traffic counts for the surrounding road network, querying the Census API for ACS tract-level demographics, downloading the zoning layer, and mapping competitors from the business license database is work that takes multiple hours per site even for a researcher who knows exactly where to look.
CoStar's value proposition is not that it holds exclusive data. It is that it eliminates the assembly cost. When a broker or analyst needs to evaluate 20 potential sites in a week, the time savings of a single-platform workflow are real and significant.
For investors who evaluate fewer sites, or who are building a repeatable research workflow, the public data route is entirely viable. The economics change: your constraint is analyst time rather than software license cost.
How PropIntel Approaches Commercial Site Evaluation
PropIntel's data foundation for commercial properties is county assessor parcel data — the same underlying source that CoStar displays for property details. For commercial parcels, this includes building size, lot size, property type code, assessor value breakdown (land vs. improvement), and recorded sale history where the state is a disclosure state.
PropIntel's AI scoring layer applies commercial investment criteria to the parcel data, comparable sales from county records, and available distress signals. The features page covers the commercial scoring parameters in detail.
For pricing context: PropIntel's commercial tier is designed for independent investors and smaller operators who need serious data tooling without enterprise-scale licensing costs. The pricing page outlines the current tier structure, and a direct comparison with CoStar is available for investors who want to understand where the platforms overlap and where they diverge.
The honest framing: PropIntel is not a CoStar replacement for institutional workflows that require lease comp data and broker-network connectivity. It is a serious alternative for investors who need parcel intelligence, distress screening, and AI-assisted deal evaluation — built on the same public data sources, without the enterprise price tag.