Real Estate Actor Suite

What does Real Estate Actor Suite do?

This Actor runs isolated adapters for every portal in the supplied catalog, including Zillow, Auction.com, LoopNet, Otodom, Cian, Funda, SeLoger, Immobiliare, Immowelt, ImmoScout24, Australian portals, UAE portals, UK portals, Canadian portals, Idealista, Apartments.com, Bayut, Zoopla, Realtor.com, accommodation portals, and the public-page Skip Trace adapter. Each adapter follows the same Jobright-style Crawlee and Playwright structure while keeping its own portal configuration in src/actor-configs.js.

The suite extracts only publicly visible or publicly embedded page data. It does not invent values. Missing source values are represented by non-null empty strings, zero values, or empty arrays and are listed in extractionWarnings.

Why use this Actor?

• Run one adapter, a selected group, or all configured adapters.
• Start with a configured public default URL for each portal, or override it with portal-specific URLs through actorInputs.
• Open detail pages for enrichment and retain structured JSON-LD, metadata, scoped card data, and selected JSON API responses.
• Deduplicate records per actor by canonical listing URL and identifier.
• Capture per-actor counters, block diagnostics, and failures in RUN_SUMMARY.
• Use public HTTP HTML for Centris, HotPads, Trulia, Zillow, Rightmove and Trip.com; use the browser for JavaScript-rendered portals such as Agoda. Embedded JSON and property-specific selectors supply listing details.
• Access the dataset through the Apify API, schedule runs, and monitor output from the Console.

Standalone per-portal Actors

Each catalog entry is also available as an independent Jobright-style Actor under . Every child folder contains its own .actor metadata and schemas, src runtime, storage results and diagnostics, output exports, README.md, Dockerfile, package files, and dataset validator. Run one child folder with apify run; regenerate the complete layout with node scripts/scaffold-standalone-actors.mjs.

What data can it extract?

Each record includes the originating actor, portal, canonical URL, listing ID, title, description, transaction type, property type, price, address, bedrooms, bathrooms, area, year built, agent/company information, public contact fields, photos, amenities, raw structured data, quality flags, and extraction warnings.

How to scrape property portals

1. Select the required actor names in the Input tab, or keep all to run every adapter.
2. The shared startUrls input already contains one public default URL per configured portal. Replace those URLs when you need a different location/search, or provide portal-specific URLs under actorInputs.
3. Set maxItemsPerActor, maxPagesPerActor, and includeDetails.
4. Run the Actor and inspect the Detailed Listings from All Actors dataset view.
5. Use the actorName and sourcePortal fields to split results by portal.

For local development, install dependencies with npm ci, validate the schema with npm run validate, and run with apify run. The testMode input runs a fast offline fixture smoke test for every selected adapter.

Input

See the Input tab for all configuration options. For different URLs per actor, use this shape:

{
  "actors": ["zillow-scrape-address-url-zpid", "realestate-com-au-scraper"],
  "actorInputs": {
    "zillow-scrape-address-url-zpid": {
      "startUrls": [{ "url": "https://www.zillow.com/" }]
    },
    "realestate-com-au-scraper": {
      "startUrls": [{ "url": "https://www.realestate.com.au/buy" }]
    }
  },
  "maxItemsPerActor": 10,
  "includeDetails": true,
  "includeRawData": true,
  "proxyEnabled": false,
  "maxRequestRetries": 3,
  "sameDomainDelaySecs": 3
}

proxyEnabled defaults to false, using the verified public HTTP route where configured. Set it to true to opt into the browser with an authorized Apify proxy or user-supplied proxyUrls inside proxyConfiguration. Proxy access is account- and plan-dependent and may incur costs. Unavailable access is reported; CAPTCHA and authentication requirements are not solved automatically. Fixture tests do not establish live portal coverage.

The default startUrls cover 31 portal adapters. The Skip Trace adapter has no fabricated default people-search URL: it requires authorized public startUrls supplied by the user.

For the recovered actors, use scripts/live-http-input.json and scripts/live-recovery-input.json with apify run --input-file. Keep separate relative APIFY_LOCAL_STORAGE_DIR and CRAWLEE_STORAGE_DIR values to preserve earlier runs. Raw platform datasets retain extraction warnings; the delivery exporter omits unknown values and never invents missing details or inserts error rows into data.

The Skip Trace adapter intentionally requires user-supplied public URLs. It does not execute a default person lookup.

Output

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel. Every emitted row has no JSON null values. When a portal does not publish a field, the row remains valid and records the field name in extractionWarnings.

{
  "actorName": "realestate-com-au-scraper",
  "sourcePortal": "Realestate.com.au",
  "title": "Example property listing",
  "listingUrl": "https://www.realestate.com.au/property-example",
  "price": { "display": "$750,000", "value": 750000, "currency": "AUD", "period": "", "raw": "$750,000" },
  "address": { "full": "Example suburb", "street": "", "locality": "Example suburb", "region": "", "postalCode": "", "country": "AU", "latitude": 0, "longitude": 0 },
  "features": { "bedrooms": 3, "bathrooms": 2, "area": { "display": "", "value": 0, "unit": "" }, "yearBuilt": 0, "floor": "", "totalFloors": "", "parking": "", "furnished": "" },
  "extractionWarnings": ["area", "yearBuilt", "photos"]
}

Cost, compliance, and support

Compute-unit and proxy costs depend on the number of portals, pages, detail requests, retries, and proxy type. Start with a small maxItemsPerActor value and increase it after checking coverage and blocking. Respect each portal's terms, robots directives, rate limits, and applicable privacy laws. The Skip Trace adapter can encounter personal data; use it only with a legitimate, documented purpose and appropriate authorization. Public data is not automatically unrestricted data.

For troubleshooting, inspect RUN_SUMMARY, per-actor ACTOR_SUMMARY_<actor-name> keys, and blocked-page diagnostics. Use the Issues tab for selector changes or portal-specific feedback.