How to run US Property Records Export & Owner Lookup (County Assessor)

Investors building an absentee-owner mailing list, property managers looking for landlords, and data teams that need official owner and value fields all hit the same wall: the assessor roll is public, but every county publishes it its own way. Maricopa serves an ArcGIS layer, Cook a Socrata catalogue split across four datasets you have to join on the PIN, Philadelphia a Carto SQL endpoint, Harris a 212 MB zip, and Florida 67 county files on a SharePoint site. For one address, the county’s own search page is the right tool. For one ZIP code of 15,926 parcels (85323 in Avondale, Arizona) it is a spreadsheet job measured in days, and the next county starts from zero.

What you’ll need

Nothing but an Apify account. The actor talks to County & state assessor rolls (28 feeds) directly.

No API key, no county account, and no proxy: all 28 feeds are published by the assessor or the state for machine use, and the proxy setting ships off. To see what is covered before you spend anything, tick listCoverage and run it: one free row per feed, 525 counties in 22 states as of 28 October 2026 (Massachusetts was added that day), each row listing the fields that feed publishes.

Step 1 — Open the actor

Open US Property Records Export & Owner Lookup (County Assessor) on the Apify Store and press Try for free. The actor’s own page on this site lists its live health and pricing.

Step 2 — Fill in the input

Field	Type	Required	What it does
areas	array	no	One per line. A state (‘FL’ or ‘Florida’), a county (‘Miami-Dade, FL’, ‘Cook County, Illinois’, ‘IL/Cook’) or a 5-digit ZIP code (‘85323’). A ZIP becomes its county plus a ZIP… (default [])
area	string	no	Same as one line of ‘Areas to export’; kept for saved tasks and API callers that send one value.
preset	string	no	A ready-made filter for common lead lists. Any filter you set below overrides the preset’s value for that filter. Absentee presets leave out government owners. (default none)
countOnly	boolean	no	Don’t export: return, per area, how many properties match and the estimated cost of exporting them. Free. Counts are exact for ZIP/city/value filters; owner-side filters…
maxResultsPerArea	integer	no	Cap on rows returned for EACH area line (not per run), so adding areas never cuts off the later ones. 0 = no cap: the whole filtered area, bounded only by your run’s max charge… (default 1000)
zipCodes	array	no	Only properties in these ZIP codes. (default [])
cities	array	no	Only properties whose situs (property) city is one of these, e.g. ‘OAK LAWN’. Case does not matter. (default [])
propertyUse	array	no	Words to match in the county’s property-use description (‘single family’, ‘condo’, ‘vacant’, ‘commercial’, ‘multi’) or a class-code prefix. Matches if ANY entry matches. (default [])
ownerTypes	array	no	Keep only these owner types, one per line: individual, company (LLC, Inc, LP, …), trust, government. Classified from the owner name. (default [])
absentee	array	no	Keep only these occupancy tiers, one per line: out_of_state, out_of_city, same_city_diff_address, owner_occupied, unknown. Compares the owner’s mailing address with the… (default [])
minValue	integer	no	Market value, or assessed value where the county publishes no market value.
maxValue	integer	no	Market value, or assessed value where the county publishes no market value.
soldAfter	string	no	YYYY-MM-DD. Keeps properties whose last recorded sale is on or after this date (month precision where the county records only the month).
soldBefore	string	no	YYYY-MM-DD. Keeps properties whose last recorded sale is before this date.
minYearBuilt	integer	no	Year built, where the county publishes it.
maxYearBuilt	integer	no	Year built, where the county publishes it.
lookups	array	no	E.g. ‘1001 Preston St, Houston, TX 77002’, ‘IL/Cook/17-09-100-001-0000’, ‘FL/Miami-Dade/ACME HOLDINGS LLC’. No limit on how many per run. (default [])
addresses	array	no	Street addresses with city, state and preferably ZIP. Same as ‘Lookups’, but always read as addresses. (default [])
parcelLookups	array	no	STATE/County/ParcelID, e.g. ‘IL/Cook/17-09-100-001-0000’. A bare ID works with the State and County hints below. (default [])
ownerLookups	array	no	STATE/Owner Name or STATE/County/Owner Name. Matches names that START with what you type. Returns every parcel found plus a free portfolio summary row. (default [])
address	string	no	One address; kept for API callers that send one value.
parcelLookup	string	no	One parcel lookup.
ownerLookup	string	no	One owner lookup.
stateHint	string	no	Two-letter state for bare parcel IDs or owner names, e.g. ‘IL’.
countyHint	string	no	County for bare parcel IDs or addresses, e.g. ‘Cook’ (use with State hint).
maxResultsPerOwner	integer	no	Cap on properties returned for EACH owner lookup. (default 50)
includeHistory	boolean	no	Add every recorded sale (with buyer and seller where published) and prior-year values to each row. Turn off for a smaller, flatter dataset. (default True)
listCoverage	boolean	no	Return one free row per county/state source this actor reads, with the fields each one publishes. Nothing else runs.

Most of that table is filters and lookups you will not touch on a first run. Two fields decide what you get.

areas takes a state, a county or a ZIP, one per line, and a ZIP turns into its county plus a ZIP filter, so 85323 quietly means “Maricopa County, ZIP 85323”. A state reads the whole statewide roll where one exists (Florida’s is 67 county files), which is why the cap below matters.

maxResultsPerArea is a cap per line, not per run, and it defaults to 1,000. At $0.02 a row that default is $20 per area, so set it to what you want, or run with countOnly first: it answers “how many, and what would it cost” for free. For 85323 it says 15,926 parcels before any filter.

A working input:

{
  "areas": [
    "85323"
  ],
  "preset": "absentee_owners",
  "maxResultsPerArea": 100
}

Step 3 — Run it

Press Start. Rows are pushed to the dataset as they’re found, so the run log fills in as it works.

An ArcGIS or Socrata county is quick: the smoke run for this listing pulled 60 rows from four counties (Maricopa, Philadelphia, Cook and Liberty County, Florida) in 73 seconds on the Apify platform. Harris is the slow one, because its roll is a 212 MB file: the download alone took 60 seconds on Apify, and an address lookup there finished in 86 seconds.

An empty result is never silent. An area outside coverage comes back as a free not_covered row, a feed that does not answer as a free blocked row, and a filter a county cannot answer (property type where the county publishes no description, say) skips that county with a free row naming the filter, rather than handing you unfiltered rows.

Step 4 — Read the output

Each run returns a labelled table, not raw JSON:

Every row looks like this — a real row from a real run:

{
  "query": "85323",
  "query_type": "export",
  "state": "AZ",
  "county": "Maricopa",
  "county_fips": "04013",
  "parcel_id": "101-01-026",
  "situs_address": "810 S 114TH DR",
  "situs_city": "AVONDALE",
  "situs_zip": "85323",
  "owner_name": "J AND L AZ PROPERTIES LLC",
  "owner_type": "company",
  "mailing_address": "5001 142ND LN NW",
  "mailing_city": "RAMSEY",
  "mailing_state": "MN",
  "mailing_zip": "55303",
  "absentee_tier": "out_of_state",
  "absentee_owner": true,
  "market_value": 331000,
  "assessed_value": 169254,
  "tax_year": 2027,
  "year_built": 2008,
  "building_sqft": 1804,
  "lot_sqft": 6722,
  "last_sale_date": "2013-03-01",
  "last_sale_price": 128000,
  "source": "Maricopa County Assessor"
}

Field	Example value	Meaning
parcel_id	101-01-026	The county’s own parcel / account / PIN number
situs_address, situs_city, situs_zip	—	The property’s address
owner_name	J AND L AZ PROPERTIES LLC	Owner of record as the roll prints it (null where state law withholds names: California, New Jersey, and a few counties)
owner_type	company	individual, company, trust or government — classified from the name
mailing_address, mailing_city, mailing_state, mailing_zip	—	Where the owner’s tax mail goes
absentee_tier / absentee_owner	—	out_of_state, out_of_city, same_city_diff_address, owner_occupied or unknown, with absentee_basis saying why; absentee_owner is the yes/no version
market_value, assessed_value, land_value, improvement_value, taxable_value	—	The county’s values; field_notes says what each one means for that county
property_class, property_class_desc	—	The county’s use code and its description

absentee_tier is the column most people came for, and it is deliberately conservative. It compares the owner’s mailing address with the property: another state is out_of_state, another city out_of_city, the same street address owner_occupied. When the roll has no mailing address, or a county writes the township instead of the postal city (Wisconsin does), it says unknown and absentee_basis says why, rather than guessing. In a 300-row Maricopa sample, 34 owners were out of state and 24 out of city.

Expect nulls that are the county’s, not the actor’s. California and New Jersey withhold owner names, Texas and Montana publish no sale prices, and Cook publishes assessed but not market value. Every row carries field_notes saying which gaps are structural for its county.

Step 5 — Export it

Open the Dataset tab and export to CSV, JSON, or Excel — or pull the same rows from the API, which is what you want if this is going to run on a schedule.

What it costs

Two prices. An exported property row is $0.02, so the 100-row input above costs $2.00. A lookup that finds a property is $0.045. Coverage lists, countOnly previews, uncovered areas, lookups that miss and the owner-portfolio summary row are free. Apify platform usage is billed separately. A sensible first run is a countOnly pass, which is free, followed by an export capped at the number of letters you can actually send.

Where the data comes from

This reads County & state assessor rolls (28 feeds) directly. The target is 28 official county and state assessor feeds behind four readers: ArcGIS FeatureServer/MapServer query endpoints, a Socrata SODA catalogue (Cook), a Carto SQL endpoint (Philadelphia), and zipped bulk rolls (Florida DOR NAL files listed via SharePoint REST; HCAD’s 212 MB Harris file). Every night a canary runs this actor against that live source and diffs the result against a frozen fixture — what “verified” means.

The feeds are easy to reach and hard to trust. Cook County’s own dataset has a column called owner that matched the buyer of a sale since 2023 on 3 of 250 parcels, while its taxpayer column matched on 201, so this actor reads the taxpayer name. North Carolina writes 2001-01-01 as a sale date on 103,929 parcels that never sold that day, and the actor drops it rather than passing it on as a real sale.

See it used

Build an absentee owner mailing list for a ZIP code takes one Chicago ZIP from 25,087 parcels to a mailing list. Look up who owns a list of property addresses resolves four addresses in four states in one run.

For the listing side of an address rather than the county roll (the Redfin Estimate, listing history, photos), reach for Zillow & Redfin Property Data Scraper.