Zillow Scraper — Sold Homes, Rentals & Homes for Sale

Zillow Scraper turns any Zillow search into a clean spreadsheet: homes for sale, homes for rent and recently sold homes, with every filter Zillow offers, in one flat row per home — price, beds, baths, square feet, lot size, Zestimate, rent Zestimate, sold date and sold price, days on Zillow, listing agent and broker phone, MLS number, price history, photos and facts. Type a city, a ZIP code, a neighborhood or paste a Zillow URL, press Start, and download JSON, CSV or Excel — or pull the rows through the scrape.ly API into Google Sheets, n8n, Make, Zapier, a database or an AI agent.

What is Zillow Scraper?

Zillow is the largest real-estate marketplace in the United States, but its official API is closed to almost everyone and its website shows at most 820 homes per search.

Three kinds of search, one row shape:

Every row also carries the address split into street, city, state and ZIP, coordinates, home type, lot size and a one-sentence summary an AI agent or a Slack alert can read as is.

How to scrape Zillow sold homes and sold comps

Set Listing type to sold and the Actor returns recently sold homes the same way it returns active listings: one flat row per home, with soldPrice, soldDate, the full priceHistory, Zestimate at the time of the search, beds, baths, square feet, lot size and the listing agent. Narrow the window with Days on Zillow — sold within 7 days, 30 days, 90 days, up to 36 months — and the search by city, ZIP code, neighbourhood, school district or a map area drawn on Zillow.

{
  "searchTerms": ["Austin, TX"],
  "listingType": "sold",
  "daysOnZillow": "90",
  "maxResults": 2000
}

Three things worth knowing before you build comps on this data:

• The 820-result cap is bypassed. Zillow itself stops paging at about 820 homes per search; the Actor splits the map into tiles and keeps going, so a whole city of sold homes comes back in one run.
• Non-disclosure states. In twelve US states the sale price is not public record. Zillow shows what it has, and so do we: soldPrice can be null while soldDate is filled. That is the source, not a bug.
• Price history is where the story is. priceHistory carries every listing, price cut and sale event Zillow knows for that home — enough to see how long it sat and how far the asking price moved.

What data does Zillow Scraper extract?

On every row (search results):

Added with "Add full details" (or for pasted URLs, ZPIDs and addresses):

Apartment buildings (rentals): buildingName, minRent, maxRent, availableUnits, units (unit number, floor plan, beds, baths, sqft, rent, base rent, required fees, available-from date), phone, managementCompany, petPolicy, amenities, officeHours, schools, photos. Switch on One row per rental unit to get each unit as its own row instead.

Rows that fail come back as rowType: "error" with errorCode and errorMessage (for example location-not-found when Zillow does not recognise a place) — they cost nothing and nothing disappears silently.

How much does it cost to scrape Zillow?

Pricing on scrape.ly: the event prices below are what this Actor charges. Platform usage (compute, proxy, browser) is billed on top at your plan rates.

Pay per result, no subscription, no charge for errors:

So $1 buys about 770 search rows, or about 260 homes with full details. Zillow answers without a proxy, so the platform usage on top is a fraction of a cent: the 5,692-home Austin run above cost $0.04 of compute.

Set Maximum results and the run can never cost more than that number × the price — the run stops the moment it is reached. Unreachable pages, unknown locations and blocked requests are never billed.

This one has no start fee.

Bulk export: what 50,000 search results actually cost

This Actor is built for bulk jobs — put hundreds of inputs into one run, or call it from the API on a schedule. There is no fee per run and no fee per page: you pay for the rows you keep, and error rows are free. That is what decides the bill once you pull a whole market rather than a single query.

How to scrape Zillow with Zillow Scraper

1. Pick the listing type — for sale, for rent or recently sold.
2. Tell it where — one location per line: Austin, TX, 78704, Zilker, Austin, TX, Travis County, TX. Or paste Zillow search URLs straight from your browser, filters and all.
3. Add filters if you want — price, beds, baths, home types, square feet, lot, year built, listed within, seller type (FSBO, foreclosure, auction, new construction…), keywords, must-have features, HOA cap, parking.
4. Set Maximum results — this is your cost cap.
5. Start. Rows appear in the dataset as they are found; download them as JSON, CSV, Excel or HTML, or read them through the API.

<img src="https://actor-media-fw21.netlify.app/zillow-scraper/zillow-scraper-input-form.png" alt="Zillow Scraper input form: listing type, locations, Zillow search URLs, property URLs, filters and maximum results" width="900">

⬇️ Input

The simplest useful input — 100 homes for sale in one ZIP code:

{
    "listingType": "forSale",
    "locations": ["78704"],
    "maxItems": 100
}

Investor search — owner-posted and foreclosure houses under $400k with a pool, listed in the last 30 days, with full details and the agent's phone:

{
    "listingType": "forSale",
    "locations": ["Phoenix, AZ", "Mesa, AZ"],
    "sellerTypes": ["ownerPosted", "foreclosure", "auction"],
    "homeTypes": ["house"],
    "priceMax": 400000,
    "features": ["pool"],
    "daysOnZillow": "30",
    "includeDetails": true,
    "maxItems": 500
}

Sold comps for an appraisal — houses sold in the last 90 days within a neighborhood:

{
    "listingType": "sold",
    "locations": ["Zilker, Austin, TX"],
    "homeTypes": ["house"],
    "daysOnZillow": "90",
    "sort": "newest",
    "maxItems": 0
}

Rentals, one row per unit, pets allowed:

{
    "listingType": "forRent",
    "locations": ["Denver, CO"],
    "bedsMin": 2,
    "priceMax": 2500,
    "features": ["petsAllowed", "inUnitLaundry"],
    "oneRowPerUnit": true,
    "maxItems": 300
}

Homes you already know — a ZPID, a street address, a listing URL and an apartment building:

{
    "propertyUrls": [
        "331627751",
        "15000 Upland Willow Rd, Austin, TX 78724",
        "https://www.zillow.com/homedetails/1208-Kinney-Ave-Austin-TX-78704/29324051_zpid/",
        "https://www.zillow.com/apartments/austin-tx/bellrock-la-frontera/CtqCKn/"
    ]
}

Every input field

⬆️ Output

One row per home. A search row with full details:

{
    "zpid": "29325047",
    "url": "https://www.zillow.com/homedetails/1113-Bluebonnet-Ln-Austin-TX-78704/29325047_zpid/",
    "listingType": "forSale",
    "status": "FOR_SALE",
    "statusText": "Active",
    "price": 1067000,
    "pricePerSqft": 499,
    "address": "1113 Bluebonnet Ln, Austin, TX 78704",
    "street": "1113 Bluebonnet Ln",
    "city": "Austin",
    "state": "TX",
    "zip": "78704",
    "latitude": 30.2466,
    "longitude": -97.7676,
    "beds": 4,
    "baths": 2,
    "sqft": 2137,
    "lotSizeSqft": 8189,
    "homeType": "house",
    "yearBuilt": 1959,
    "zestimate": 1052300,
    "rentZestimate": 4870,
    "daysOnZillow": 65,
    "isFsbo": false,
    "isNewConstruction": false,
    "isForeclosure": false,
    "isPending": false,
    "brokerName": "Keller Williams Realty",
    "agentName": "Andrew Allen",
    "agentPhone": "(512) 346-3550",
    "mlsId": "2075799",
    "hoaFee": null,
    "heating": "Central",
    "cooling": "Central Air",
    "parking": ["Attached"],
    "description": "Charming mid-century home on a corner lot in Zilker…",
    "highlights": ["Storage shed", "Paver patio", "Two fireplaces", "Huge screened-in porch"],
    "priceHistory": [
        { "date": "2026-07-10", "price": 1067000 },
        { "date": "2026-05-02", "price": 1099000 }
    ],
    "photoCount": 30,
    "photos": ["https://photos.zillowstatic.com/fp/…-cc_ft_1536.jpg", "…"],
    "facts": { "propertySubType": ["Single Family Residence"], "parkingCapacity": 2, "lotSize": "8,189 sqft" },
    "searchLocation": "Zilker, Austin, TX",
    "summary": "1113 Bluebonnet Ln, Austin, TX 78704: 4 bd 2 ba 2,137 sqft house $1,067,000; 65 day(s) on Zillow; listed by Andrew Allen ((512) 346-3550).",
    "scrapedAt": "2026-09-13T15:00:12Z"
}

A sold home:

{
    "zpid": "7946659",
    "listingType": "sold",
    "status": "SOLD",
    "price": 550000,
    "soldPrice": 550000,
    "soldDate": "2026-09-11",
    "address": "4037 W Kelton Ave, Phoenix, AZ 85053",
    "beds": 4,
    "baths": 2,
    "sqft": 2700,
    "homeType": "house",
    "zestimate": 543100,
    "rentZestimate": 2650,
    "brokerName": "Arizona Premier Realty Homes & Land, LLC",
    "summary": "4037 W Kelton Ave, Phoenix, AZ 85053: 4 bd 2 ba 2,700 sqft house sold for $550,000 on 2026-09-11; Zestimate $543,100."
}

An apartment unit (with One row per rental unit):

{
    "listingType": "forRent",
    "isBuilding": true,
    "buildingName": "Chimney Park Apartments",
    "unitNumber": "Unit 310",
    "floorPlan": "A1",
    "beds": 1,
    "baths": 1,
    "sqft": 450,
    "price": 1049,
    "baseRent": 949,
    "requiredMonthlyFees": 100,
    "availableFrom": "2026-11-10",
    "address": "2406 Bluebonnet Ln, Austin, TX 78704",
    "phone": "(512) 555-0134",
    "petPolicy": ["Cats", "Small dogs"],
    "minRent": 1049,
    "maxRent": 1890,
    "availableUnits": 14,
    "summary": "Chimney Park Apartments, Unit 310: 1 bd 1 ba 450 sqft $1,049/mo, available 2026-11-10 at 2406 Bluebonnet Ln, Austin, TX 78704."
}

<img src="https://actor-media-fw21.netlify.app/zillow-scraper/zillow-scraper-output-table.png" alt="Zillow Scraper output table: address, listing type, price, sold date, beds, baths, sqft, home type, Zestimate, days on Zillow, agent and phone" width="900">

Use cases for Zillow data

Real-estate investors and wholesalers

Run the FSBO + foreclosure + auction filter over your target metros every morning with Monitoring mode on: each run returns only homes that are new or whose price or status changed since yesterday. Pipe the rows into a sheet or a CRM and call the agentPhone on the ones that match your buy box. Sold comps for the same ZIP come from the same Actor with listingType: "sold".

Agents and brokerages

Build a list of every active listing in your farm area with the listing agent's name, phone and brokerage — recruiting, co-broking and expired-listing outreach start from that table. Track days on Zillow and price cuts to time your calls.

Appraisers, analysts and PropTech teams

Pull all recently sold homes in a county with sold date, sold price, beds, baths, sqft and lot size, and compute price per square foot by neighborhood. Schedule the run weekly and append to a warehouse; the flat schema and stable field names make the append trivial.

Property managers and rental platforms

Rent comps by ZIP, building by building, unit by unit: floor plan, sqft, base rent, required fees and availability date, plus the leasing-office phone and pet policy. Filter apartments by pets allowed or in-unit laundry the way a tenant would.

AI agents, RAG and research

Every row has a summary sentence and a stable id, so an agent can search a city, reason over prices and call the Actor again for details on the homes it picked. The Actor is listed for the scrape.ly MCP server and for agentic payments.

Integrations

• API — POST with the input JSON returns the rows directly; for longer runs start the run, wait for SUCCEEDED, then read datasets/{id}/items?format=csv`.
• Python / JavaScript — scrapely-client: client.actor("freshowl21/zillow-scraper").call(run_input={...}), then client.dataset(run["defaultDatasetId"]).list_items().
• Google Sheets — scrape.ly's Google Sheets integration, or the CSV export link, or the scrape.ly node in n8n/Make that appends dataset rows to a sheet. The rows are flat, so every column maps without transformation; the few nested fields (priceHistory, photos, units, facts) arrive as JSON text.
• n8n, Make, Zapier — use the scrape.ly app / node, pick this Actor, paste the input JSON, and map the dataset items downstream. Set Maximum results so a big city does not overflow a scenario's limits; for runs over 5 minutes use the "wait for finish" option or a webhook rather than a plain HTTP request.
• Webhooks — an Actor-run webhook on ACTOR.RUN.SUCCEEDED posts the dataset link to your endpoint or Slack.
• Scheduling — scrape.ly Schedules run the same input daily or hourly; combined with Monitoring mode this is a change feed of a market.
• MCP — connect the scrape.ly MCP server to Claude, Cursor or any MCP client and ask for Zillow data in plain language; the Actor's input schema is the tool description.

Zillow's result cap and how it is handled

Zillow shows at most 20 pages of 41 homes for any single search, so a search for a big city normally stops at 820 homes no matter what. With Bypass Zillow's 820-result cap on (the default) the Actor reads the total first and, when it is above the cap, splits the map into four quadrants, then again, until every quadrant is under the cap, and merges the results with duplicates removed. The city boundary is kept — a quadrant search still respects the region, so homes outside the city limits do not creep in. The whole of Austin (5,692 homes for sale) took 153 requests and 4 minutes.

Sold prices and non-disclosure states

Twelve US states (Texas, Utah, Idaho, Montana, Wyoming, North Dakota, Kansas, Missouri, Mississippi, Louisiana, Alaska, New Mexico) do not make sale prices public. Zillow shows a sold date but no sold price there, and so does this Actor — soldPrice is null rather than a guess. In every other state both fields are filled.

Data sent outside scrape.ly

To collect the data, this Actor sends requests through a residential proxy provider (DataImpulse) and, as a last resort, a backup proxy service, which receive the Zillow page addresses being requested. Only what is needed for the request is sent; no scrape.ly credentials, account data or results are shared.

❓ FAQ

Is it legal to scrape Zillow?

The Actor reads only what Zillow shows to any visitor without logging in — public listing pages — and stores no personal data beyond what the listing itself publishes (the listing agent's business contact). Scraping publicly available data is generally lawful, but how you use it is your responsibility: respect Zillow's terms where they apply to you, fair-housing rules and privacy law in your jurisdiction, and consult a lawyer for commercial use. Do not use the data to contact homeowners against their wishes.

Do I need a Zillow account, API key, cookies or a proxy?

No. Nothing to log into, nothing to paste, nothing to configure. If a very large run ever starts seeing blocked error rows, switch on scrape.ly residential proxy in the advanced section — but this has not been necessary in testing.

Why does my search return more homes than Zillow shows me?

Because of the cap bypass. Zillow shows at most 820 per search; the Actor splits the map and collects them all. Turn bypassResultCap off if you want exactly what one Zillow page set shows.

Why is soldPrice empty for Texas?

Texas is a non-disclosure state; Zillow does not publish sale prices there. The sold date is still returned. See the section above.

Can I search by address, ZPID or a list of URLs?

Yes — put them in Property URLs, ZPIDs or addresses, one per line. Each returns a fully detailed row. Mixed lists are fine. An address still finds its home when your list uses a postal city or ZIP that Zillow files differently — the house number and street decide.

Can I use the filters from a Zillow URL I already have?

Yes. Paste the URL into Zillow search URLs; its filters are read from the address bar and any filter you set in the form is added on top.

How do I get only new listings every day?

Set Monitoring mode — store name to any name, schedule the run daily. The first run returns everything; later runs return only homes that are new or whose price or status changed, each marked changeType: "new" or "changed" (with previousPrice).

How fast is it?

About 41 homes per request for search, one request per home for details. 100 search rows take a few seconds; 100 fully detailed homes about 15–30 seconds at the default concurrency.

Does it get tax history and school ratings?

Not yet. Zillow loads those two blocks through a separate, signed request that anonymous visitors cannot reproduce reliably, and I would rather leave a field out than fill it with guesses. Everything else on the listing page is included. If you need them, open an issue — it is on the list.

Can I get one row per apartment unit?

Yes — oneRowPerUnit: true. Otherwise a building is one row with its units nested under units.

Can I use Zillow Scraper with the scrape.ly API, Python, n8n or Make?

Yes, see Integrations above. The Actor follows the standard scrape.ly run → dataset flow, so every scrape.ly integration works unchanged.

Can I run it through an MCP server or let an AI agent call it?

Yes. It is available through the scrape.ly MCP server, the input fields are described for agents, and it is enabled for agentic payments.