Listing · PROPRAVEN_PARCEL_INTELLIGENCE
PropRaven Parcel Intelligence
Every U.S. parcel, every signal, one row.
Engineers · Data scientists · App buildersRefreshed monthly
Description
Parcel Intelligence is the marquee wide table — every U.S. parcel pre-joined to every enrichment we ship, on a single row. It replaces 6+ vendor data feeds with one Snowflake share.
If you're integrating property intelligence into an application or building features against the data, this is the right access mode: one query, one row per parcel, all 583 fields available without joins.
What's included
- Identity & geometry: APN, normalized APN, lat/lon, geometry polygon, lot size, county/CBSA/state.
- Property attributes: year built, building area, beds, baths, units, owner-occupancy flag, MISMO-normalized property type (96.4% classified).
- Owner intelligence: 27.56M canonical entities (72.8% coverage), entity type (LLC / CORP / TRUST / GOVT / INDIVIDUAL / ...), state of formation.
- Tax & assessments: latest tax year + multi-year history, assessed values (land/improvement/total), tax amount, exemptions.
- Permits: building-permit count, recent permit activity, contractor names, business licenses, NAICS codes.
- Deeds & sales: last sale date/price, prior sale, deed count, grantor/grantee.
- AVM & valuation: PropRaven AVM with confidence (87.6% coverage).
- Risk vector: FEMA NRI overall + per-peril (94.9% rated), NFIP flood claims, NOAA storms, USGS seismic, CDC SVI, wildfire score, climate risk tier.
- Demographics: ACS tract attributes, IRS migration, HMDA loan activity, CDC PLACES health.
- Crime: FBI/NIBRS-derived rates and concentration.
- Federal overlays: HUD HMDA / FMR / LIHTC QCT, FDIC SOD branches, Justice40, EPA ECHO, FEMA disaster history.
Use cases
- CRM enrichment: hydrate a customer record with the full property profile in a single SELECT.
- Pricing models: train AVM-supplement or insurance-pricing models against pre-resolved features.
- Lead scoring: filter to your buyer cohort with one WHERE clause across 583 dimensions.
- Embedded apps: power "property dashboard" features without managing 6 vendor pipelines.
Sample query
SELECT
PARCEL_ID,
PROPERTY_ADDRESS,
PROPERTY_TYPE_CLASS,
YEAR_BUILT,
TOTAL_ASSESSED_VALUE,
PROPRAVEN_AVM,
NRI_RISK_RATING,
FLOOD_ZONE,
OWNER_NAME,
OWNER_ENTITY_TYPE
FROM PROPRAVEN_PARCEL_INTELLIGENCE.CURATED.V_MARKETPLACE_NATIONAL
WHERE STATE_FIPS = '06'
AND PROPERTY_TYPE_CLASS = 'SFR'
AND NRI_WILDFIRE_RATING IN ('Very High', 'Relatively High')
AND PROPRAVEN_AVM > 750000
LIMIT 100;
Refresh cadence
Refreshed monthly. Source data inherits its publishing cadence — county assessor records refresh monthly to annually, FEMA NFHL/NRI annually, FBI/NIBRS quarterly, Census TIGER 2025 and ACS annually, FDIC SOD annually, HUD HMDA/FMR/LIHTC QCT annually, CEQ Justice40 quarterly. Per-row _LOADED_AT and _SOURCE_AS_OF columns expose the actual freshness for any row in your query result.
Upgrade path
Parcel Intelligence is one of three access modes for the same underlying truth:
- Parcel Intelligence (this listing) — wide table, one row per parcel, single-join workflows.
- Ontology — graph shape with canonical entities + edges; better for traversal queries ("every parcel owned by entities controlled by X").
- Vertical bundles (Lending / Insurance / Banking / SFR / CRE / GovTech) — pre-filtered cohorts for specific use cases, often cheaper.
Subscribers to Parcel Intelligence can layer in the Ontology share or any vertical bundle without redundant data ingestion.
Try before you buy
Subscribe to PropRaven Preview for a free 10K-parcel Mecklenburg County NC sample with the same schema. Once you've validated your queries against the sample, subscribe to PROPRAVEN_PARCEL_INTELLIGENCE for full national data.
Questions? hello@propraven.com — or browse the data dictionary and sample queries.