Zillow Scraper — Search Results and Full Listing Details

$0.80 per 1,000 properties ($0.0008 each), plus $0.001 each time a run starts — a flat rate on every plan, with search results and full listing details coming from this one Actor rather than from two separate ones you have to stitch together.

Scrape any Zillow search or listing URL and get the full picture — price, beds, baths, sqft, address, photos, price history, Zestimate and school ratings. 80 properties in under 20 seconds.

Feed it a city search and get every result; feed it a listing and get everything Zillow shows. No login, no API key, no browser to babysit.

Who uses this

• Investors screening entire markets in seconds instead of clicking through listings.
• Agents building comps for a valuation in one run.
• PropTech teams powering valuation and lead-scoring models.
• Analysts tracking price movements across a ZIP or metro over time.

What you get for $0.0008 a property

	This Actor
Price per property	$0.0008, flat — no volume tiers, no plan gates
Run-start fee	$0.001 per run, whatever the run returns
Search and detail pages	Both, one Actor — search cards and /homedetails/ records
Price history, tax history + schools	Included, no surcharge, via includeHistory / includeSchools
Photos	Included, via includePhotoUrls
Login or API key	Never
Charged on failure	No — blocked, not-found and no-result rows are uncharged

Frequently asked questions

Do I need a Zillow account or API key? No. Paste a public URL.

Can I scrape a whole city? Yes — pass a Zillow search URL and set maxSearchResultsPerUrl.

Does it return price history and schools? Yes, with includeHistory and includeSchools.

Can I get photos? Yes, via includePhotoUrls.

Am I charged when a listing cannot be read? Not the per-result fee — only the $0.001 run-start fee. Failures are returned as uncharged diagnostics.

What does a run cost? $0.0008 per property plus $0.001 per run start. 1,000 properties is $0.80 in result fees; a single run of five search URLs at 50 cards each returns up to 250 properties for $0.201 all-in. Price history, tax history, schools and photos do not change the per-property price.

The actor does not require a Zillow login, an API key, or a private data source. It reads the public page over plain HTTP — no browser to launch, which is why runs are fast and cheap — and can retry a blocked page once through a rotating United States datacenter proxy when you enable that option.

Input

{
  "searchUrls": ["https://www.zillow.com/austin-tx/"],
  "maxSearchResultsPerUrl": 10,
  "propertyUrls": [
    "https://www.zillow.com/homedetails/7300-Quicksilver-Cir-Austin-TX-78744/29508680_zpid/"
  ],
  "includePhotoUrls": true,
  "includeHistory": true,
  "includeSchools": true,
  "enableResidentialFallback": false
}

searchUrls accepts up to five public Zillow result pages. Choose the location and any public filters in Zillow first, then paste the browser URL. Each page is loaded once and the actor emits at most maxSearchResultsPerUrl (1-50, default 10) deduplicated listing cards. It does not call undocumented or private Zillow APIs, expand maps, or paginate.

propertyUrls accepts up to 10 canonical public Zillow /homedetails/ URLs. The actor removes tracking parameters and deduplicates identical URLs. Supplying neither input returns one clearly labelled _sample: true row and does not charge a result event.

Output

Every genuine public property record has ok: true and includes:

• zpid, url, address, streetAddress, city, state, zipcode, and county
• homeStatus, homeType, price, bedrooms, bathrooms, livingArea, lotSize, and yearBuilt
• zestimate, rentZestimate, daysOnZillow, description, broker/agent attribution when public
• public photos, priceHistory, taxHistory, and schools when enabled and rendered by Zillow
• transport showing direct_http on the first pass, or datacenter_http / customer_http / isp_http when the optional retry produced the row

Zillow omits fields for some properties and statuses. Such fields are null or an empty array; this does not make an otherwise public record invalid.

Search results have type: "search_listing" and include the public card fields Zillow renders: zpid, url, address, price, beds, baths, living area, Zestimate, status, photo URL when available, searchUrl, and rank. Search cards do not claim detail-only fields such as history or school data.

Charging and proxy behavior

The actor charges the property event only after it pushes a genuine ok: true property or search-listing record. It never charges sample, invalid-input, blocked, rate-limited, not-found, or no-results diagnostic rows.

Direct traffic is always attempted first. When enableResidentialFallback is off, a Zillow block returns an uncharged BLOCKED row. When it is on, the actor performs at most one proxy retry per blocked property or result page.

That retry never uses a metered proxy group. Apify Residential and SERP are billed per gigabyte, so a request for either group is replaced with rotating datacenter addresses and the substitution is reported back on the affected rows in a proxyNotice field. If you supply your own proxy servers in proxyConfiguration.proxyUrls they are used exactly as given, in the order given, and are never substituted.

Responsible use

This actor reads public listing pages only. It does not use login credentials, CAPTCHA solving, hidden endpoints, session cookies, or fingerprint masking. Use public data responsibly and ensure your workflow complies with Zillow's terms and applicable privacy and data-use laws.

Billing

Two charges apply: a $0.001 run-start fee each time a run begins, and $0.0008 per property returned — $0.80 per 1,000. Flat rate, no volume tiers, no plan gates. Samples, diagnostics, duplicates, blocked look-ups and no-result runs never incur the per-result fee — a run that returns nothing costs only the $0.001 start fee.