Redfin Scraper

What does Redfin Scraper do? It extracts for-sale listings from Redfin — listing price, beds/baths, sqft, address, neighborhood, HOA, MLS status, agent info, photos and price history — and returns them as clean JSON rows. Built for US realtors, investors, proptech and AVM/lead-gen teams who need Redfin data in a pipeline instead of a browser. Running on the Apify platform you get API access, scheduling, proxy rotation and monitoring out of the box.

Why use Redfin Scraper?

• Investor deal screening: filter listings by price, beds and location.
• AVM / comp pipelines: price history plus Redfin Estimate history per property.
• Lead generation: listing agents and brokers per active listing.
• Market monitoring: schedule daily runs per city, county or ZIP.

How it works (the bypass — no browser needed)

Redfin's homepage is guarded by AWS WAF + CloudFront (a plain request returns a captcha challenge), so this actor never scrapes HTML. Instead it talks to the same internal Stingray JSON API the Redfin frontend uses:

1. Enumeration — GET https://www.redfin.com/stingray/api/gis with region_id + region_type (1 = neighborhood, 2 = ZIP, 5 = county, 6 = city). Returns up to ~350 listings per page with price, beds/baths, sqft, address, MLS id/status, agent, photo metadata, lat/lng and the propertyId/listingId pair. Verified working over plain HTTP while the homepage served a captcha to the same IP — this endpoint is the core bypass.
2. Optional add-ons — pick a detailMode to add fast detail endpoints, each charged as its own event:
   • Historical Data → stingray/api/home/details/avmHistoricalData (Redfin Estimate value history + sale history)
   • Neighborhood Scores → stingray/api/home/details/neighborhoodStats/statsInfo (walk/bike/transit scores)
   • Full → both. Summary → GIS data only, no add-on requests or charges.

Note: aboveTheFold / belowTheFold are no longer called — they are reliably blocked by Redfin's WAF for datacenter/anonymous IPs, and the unique data they carry (description and HOA) is already present in the fast GIS response. This keeps the run fast and 403-free. 3. Related public techniques this builds on: the reteps/redfin Python wrapper's endpoint map and the community gis-search/gis-csv parameter docs. No Algolia/Meilisearch backend or embedded __NEXT_DATA__ blob exists on Redfin — the Stingray API is the structured surface.

The GIS response also carries HOA, full description (listingRemarks), soldDate, timeOnRedfin, stories, tour flags and more — all extracted for free with zero extra requests. No Algolia/Meilisearch backend or embedded __NEXT_DATA__ blob exists on Redfin — the Stingray API is the structured surface.

How to use Redfin Scraper

1. Paste one or more Redfin search URLs in Redfin search or property URLs — city, county, or neighborhood pages like https://www.redfin.com/city/29439/WA/Bothell, or individual property pages. The region id is embedded in the URL, so there is no lookup round-trip and the run is fast and reliable. ZIP codes and free-text place names are intentionally NOT accepted (Redfin's WAF blocks the page that would resolve them) — paste URLs instead.
2. Set Max listings (free accounts get a 10-listing preview automatically).
3. Choose Details to include — Summary (fast, free), Historical Data, Neighborhood Scores, or Full (both). Historical Data and Neighborhood Scores are optional add-ons billed at their own per-listing price.
4. Run and download the dataset as JSON, CSV or Excel.

Input

Field	Description
startUrls	Redfin city/county/neighborhood URLs or property pages (required) — e.g. https://www.redfin.com/city/29439/WA/Bothell. Paste one or more. The region id is in the URL, so no lookup is needed and runs are fast. ZIPs/free-text names are not supported (Redfin's WAF blocks the resolver).
maxItems	Max listings to return (default 50, hard max 1000).
detailMode	summary (default) · historical · neighborhood · full. Picks the detail add-ons per listing. historical = Redfin Estimate value history + sales; neighborhood = walk/transit/bike scores; full = both. Each add-on is charged per listing only when it returns data.
minPrice / maxPrice / minBeds	Optional filters passed to the search API.
proxyConfiguration	Native Apify proxy editor: default ON with Apify Proxy AUTO; pin a residential group (e.g. RESIDENTIAL) or country code to control the tier.
maxRequestsPerCrawl	Hard request cap for loop safety (default 600).

Example input:

{
  "startUrls": [{ "url": "https://www.redfin.com/city/29439/WA/Bothell" }],
  "maxItems": 20,
  "detailMode": "full",
  "proxyConfiguration": { "useApifyProxy": true }
}

Output

Each row is one listing. You can download the dataset in JSON, CSV, HTML or Excel.

{
  "url": "https://www.redfin.com/WA/Bothell/18108-19th-Dr-SE-98012/home/2791097",
  "address": "18108 19th Dr SE",
  "city": "Bothell",
  "state": "WA",
  "zip": "98012",
  "price": 679950,
  "beds": 3,
  "baths": 2,
  "sqft": 1041,
  "lotSize": 6534,
  "pricePerSqft": 653,
  "mlsId": "2566592",
  "mlsStatus": "Active",
  "daysOnMarket": 26,
  "listingAgent": "Marie Hartung",
  "listingBroker": "Redfin",
  "neighborhood": "…",
  "hoa": "…",
  "photos": ["https://…jpg"],
  "photoCount": 24,
  "priceHistory": [{ "price": 679950, "date": "…" }],
  "scrapedAt": "2026-09-06T…"
}

Main data fields (all from the fast GIS request): url, address, city, state, zip, latitude, longitude, price, beds, baths, sqft, lotSize, pricePerSqft, propertyType, yearBuilt, mlsId, mlsStatus, daysOnMarket, hoa, description, soldDate, timeOnRedfin, originalTimeOnRedfin, stories, fullBaths, partialBaths, isHot, isRedfin, hasVirtualTour, listingAgent, listingBroker, sellingBroker, photoCount. With detailMode: historical (or full): priceHistory, avmHistory. With detailMode: neighborhood (or full): walkScore, transitScore, bikeScore.

How much does it cost to scrape Redfin?

Per-listing (pay-per-event): the base result is $0.003 (~$3 per 1,000 listings). Optional add-ons are charged only when their data is returned:

Event	What you get	Price
result	One listing, GIS summary (price, beds, baths, sqft, address, HOA, description, agent, photos count, …)	$0.003
historicalData	Redfin Estimate value history + sale history	+$0.002
neighborhoodScores	Walk / transit / bike scores	+$0.001

A Full run (both add-ons) comes to $0.006 per listing ($6 per 1,000). A Summary run is $0.003 per listing ($3 per 1,000). Free-plan users get a 10-listing preview per run. One search page = 1 request; each add-on listing = 1 extra request. Compute and proxy usage are a few cents per ~1,000 listings at low concurrency.

Tips

• Prefer city or neighborhood regions over huge counties to keep result sets focused.
• Use minPrice/maxPrice/minBeds to cut detail-request volume before it happens.
• For thousands of listings, run one actor task per region and merge datasets.
• If detail fields come back empty, the detail IPs were likely flagged — retry later or switch proxy groups; core GIS fields (price, beds, baths, HOA, description) still flow, and you're never charged the add-on events when the data didn't come back.

FAQ, disclaimers and support

• Is this an official Redfin API? No. Redfin publishes no public API; this actor uses Redfin's own internal endpoints the same way its website does. Internal endpoints can change without notice — if a run returns errors, please report it via the Issues tab.
• Legality: scrape public listing data responsibly and review Redfin's Terms of Use for your use case. This actor is built for legitimate real-estate analytics.
• Limitations: photo/description shapes vary by listing and missing values are returned as null. Walk/transit/bike scores and price history require detailMode: historical, neighborhood, or full and depend on the detail endpoints being accessible from the current proxy IP — when they aren't, you're not charged for those add-ons. Photo URLs are not available (the endpoint that carries them is WAF-blocked), so only photoCount is returned.
• Found a bug or want bulk-search / sold-history support? Open an issue — custom builds are available.