Realtor.com Scraper: Homes for Sale, Rent & Sold

🏠 What is Realtor.com Scraper?

Realtor.com Scraper extracts property listings from Realtor.com: price, beds, baths, square footage, the full description, every photo, and the listing agent's name, email and phone, for any US city, ZIP code, county or state. It works as a Realtor.com API alternative: run it on a schedule, call it from your code, or plug it into Make, Zapier or n8n.

Type a location (Austin, TX), click Start, and download the listings in JSON, CSV or Excel. No login, nothing to set up, and it is fast: about 1,000 listings in 10 seconds. Homes for sale, for rent, sold, pending, coming soon, off market and builder plans to be built, with 31 filters and an optional detail pass that adds price history, tax history, schools, flood and fire risk and value estimates — plus two options sold apart: the property's building permit history (with a roof-permit flag) and how many people viewed the listing.

📋 What data can you extract from Realtor.com?

One item per listing, 165 fields:

Category	What you get
🏷️ Listing	Realtor.com id, MLS number, link, status (for sale, for rent, sold, pending, coming soon, off market, ready to build), who published it (an agent, a builder, a rental company), badges such as "new listing" or "price reduced"
💰 Price	asking price, price per square foot, last price cut and its date, last sale price and date — 460000 USD, 285 per sq ft
🏠 The home	property type, beds, baths, living area, lot size, year built, stories, garage spaces, HOA fee — 3 beds, 2.5 baths, 1,612 sq ft
📍 Location	street address, city, state, ZIP code, county, neighborhood, latitude and longitude, Street View link
📝 Text and photos	the agent's full description, feature lists (appliances, interior…), every photo, virtual tour, open houses, pet policy
📇 Agent and office	listing agent's name, email, phone numbers, photo and website, office with its website, broker, MLS board — (512) 637-8277
🕒 Dates	date listed, last update, days on market — what the date filters read
📈 History and estimates	detail — price history, tax history, value estimates with their past and a 12-month forecast, mortgage estimate, monthly rent estimate with its range
🏫 Schools and risk	detail — assigned schools with their rating, FEMA flood zone, flood, fire and noise scores
👀 Demand and permits	options — how many people viewed, clicked and enquired, per period (includeViewStats); building permit history and a roof-permit flag (includePermitHistory)

Every field, with an example, is listed in the Output section below.

Fields marked detail are filled when Extract details is on (read 25 listings per extra request, charged as its own event); fields marked options only when their option is on, each charged for what it delivers. Everything else — description, photos and the agent's contact details included — comes with the search results.

✅ Why use Realtor.com Listings Scraper?

• 🚀 Fast: about 1,000 listings in 10 seconds, 5 requests.
• 🏠 Type the city, not a URL: Austin, TX, 78704, Travis County or Texas — or paste a Realtor.com search URL and its filters are read.
• 🎛️ 31 filters: price, beds, baths, living area, lot, year built, garage, HOA fee, radius, property type, foreclosures, new construction, open houses, virtual tours, no pending sales, MLS listings only, pets — 27 of them applied by the site itself, so a filtered search costs less, not more.
• 📈 More than the listing card: price and tax history, school ratings, flood / fire / noise scores, value estimates with a 12-month forecast. On top, two options: the building permit history of the property (type of work, project, date, status) with a hasRoofPermit flag that spots roofing jobs at a glance — the list roofers, remodelers and insurers look for — and how many people viewed, saved and enquired about the listing over the last days. You pay for them only when they are delivered: a property with no permit on record, or a brand-new listing with no view count yet, costs nothing extra.
• 📇 The listing agent comes with the listing: name, email, phone, photo, office and broker, with no separate agent scraper to buy.
• 🔍 Past the site's 10,000-result cap: a search bigger than that is split into price bands (sold homes: sale-date ranges) on its own.
• 🔔 Monitoring built in: tick Only new listings, schedule the Actor, and each run returns (and charges) only what it has never delivered before.
• 🔌 API, scheduling, integrations (Make, Zapier, n8n, Google Sheets…), and JSON/CSV/Excel export via the Apify platform.

🚀 How to scrape Realtor.com

1. Create a free Apify account.
2. Open Realtor.com Listings Scraper and type a Location (e.g. Austin, TX).
3. Or paste your own Realtor.com URLs into Start URLs: search results pages, sold pages, rental pages, or single listing pages.
4. Pick a Listing status (for sale, for rent, sold, pending, coming soon, off market, ready to build), set the filters you need and Max listings (100 by default, 0 = no limit), then click Start.
5. Download the dataset in JSON, CSV, Excel or via API.

💰 How much does it cost to scrape Realtor.com?

This Actor uses pay per event pricing: $0.45 per 1,000 listings — plus $0.01 per run start ($1 per 100 runs). Listings read with Extract details on cost $0.69 per 1,000 instead, the extra requests included. Two options come on top, charged only for what they deliver: building permits, $1.79 per 1,000 permit rows (3 permits on a property = 3 rows; none = nothing) and view statistics, $0.89 per 1,000 listings that have them; turning either on reads the detail page, so the listing is charged as a listing with details. A single listing page pasted into startUrls is always read in full and costs the same. Higher Apify plans pay less per listing: $0.44 (Bronze), $0.43 (Silver) and $0.42 (Gold) per 1,000 — with details $0.68, $0.67 and $0.66; permit rows $1.78, $1.77 and $1.76; view statistics $0.88, $0.87 and $0.86. The Actor's own filters (listing date, excluded keywords, agent phone) are applied to every listing it reads, kept or not: $0.04 per 1,000 listings checked (filter-check), $0.03 on Bronze, $0.02 on Silver and $0.01 on Gold; a run without them pays nothing for it. Example: 20,000 listings ≈ $9.01; a daily monitor of 300 new listings ≈ $0.15 a day. Platform usage (compute, proxy) is included in the price.

⚙️ Input

{
    "location": "Austin, TX",
    "listingStatus": "for_sale",
    "maxItems": 200
}

Several locations, a cap per search, houses in a price range, only the ones not delivered before:

{
    "locations": ["Austin, TX", "78704", "Dallas, TX"],
    "propertyType": ["single_family"],
    "minPrice": 300000,
    "maxPrice": 600000,
    "minBedrooms": 3,
    "maxItemsPerQuery": 500,
    "onlyNew": true,
    "stateKey": "austin-houses"
}

Or with your own URLs:

{
    "startUrls": [
        { "url": "https://www.realtor.com/realestateandhomes-search/Austin_TX/beds-3/price-300000-600000" },
        { "url": "https://www.realtor.com/realestateandhomes-detail/7704-Fenton-Cv_Austin_TX_78736_M73634-37439" }
    ],
    "maxItems": 500
}

Give at least one place (location or locations) or one URL (startUrls). With none, the Actor runs an example instead of stopping — a search in Austin, TX, at most 20 listings — and says so in the log.

Field	Notes
location, locations	City (Austin, TX), ZIP code (78704), county or state (Texas); locations adds more places.
query, searchQueries	Keyword matched against the listing text (pool, waterfront). Every keyword is searched in every location (max 500 searches per run).
startUrls	Realtor.com search, sold or rental pages (place, price, beds, baths, sqft, lot size, home age, days on market, HOA, garage, type, radius, listing switches, sort and page are read from the URL; features, keywords and price-reduced filters are refused by name) or single listing pages.
listingStatus	for_sale, for_rent, sold, pending, coming_soon, off_market (listings that expired without a sale or a rental) or ready_to_build (plans of homes a builder will build).
propertyType	single_family, condos, townhomes, multi_family, land, mobile, farm — several at once.
minPrice, maxPrice, minBedrooms, maxBedrooms, minBathrooms, maxBathrooms	Price in US dollars (the monthly rent for rentals); half bathrooms count as 0.5.
minSqft, maxSqft, minLotSqft, maxLotSqft, minYearBuilt, maxYearBuilt, minGarage, maxHoaFee	Living area and lot in square feet, year built, garage spaces, monthly HOA fee.
radiusMiles, sortBy, soldWithinDays	Widen the search around the location (5); order the results (list_date:desc, also by price cut, open house, lot size, year built, full bathrooms, sold price for sold listings, move-in day for rentals); for sold listings, how far back to go.
foreclosureOnly, newConstructionOnly, openHousesOnly, virtualTourOnly, noHoaFeeOnly, excludePending, mlsOnly, petsAllowed	Switches applied by the site; excludePending leaves out the homes for sale already under contract (from 1 in 7 to 1 in 3, depending on the city); mlsOnly keeps the listings an MLS publishes (no builder plans, no apartments posted by rental companies); petsAllowed (cats, dogs) only applies to rentals.
priceReducedWithinDays, priceReduced	Only listings whose price dropped in the last N days, or ever (priceReducedWithinDays wins when both are set). Applied by the site.
openHouseWithinDays, openHouseFrom, openHouseTo	Only listings with an open house in the next N days (today included), or between two days (2026-11-14, 2026-11-15, that last day whole). Not together. Applied by the site.
has3DTour, withPhotoOnly	Only listings with a 3D tour or a virtual tour; only listings with at least one photo. Applied by the site.
tags, features	Only listings that carry all the site tags you give: type any tag of the site in tags (cul_de_sac, swimming_pool) or pick from the list in features. Applied by the site.
seniorCommunity	any, only or hide for the 55+ communities. Applied by the site.
mlsListingId	One listing by the number its MLS gave it (2427357); give its place too. Applied by the site.
moveInBy, rentSpecials, applyNow, builderPromotions	Rentals available by a day (2026-12-31), rentals with a special offer, rentals you can apply for online (all three for For rent only); homes for sale and builder plans with a builder promotion. Applied by the site.
soldAfter, soldBefore	For sold listings, the first and last sale day (2026-10-01, 2026-10-31, both included); with soldWithinDays the later start is used. Applied by the site.
boundingBox, polygon, centerPoint, mapAreas	Only listings inside an area of the map: a box (min / max latitude and longitude), a polygon (corners written longitude first, then latitude), a circle (a point and a radius in miles), or up to 20 such areas in mapAreas (one search each). With or without a location. Applied by the site.
street	Only listings on a street (Congress Ave) or at an address (805 Congress Ave). With or without a location. Applied by the site.
agentName, officeName	Only listings of one selling agent or of one office, by the name the site writes (Keller Williams Realty). An office needs a place (a location, a map area or a street). Applied by the site.
minElementarySchoolRating, minMiddleSchoolRating, minHighSchoolRating, schoolDistrict	Only homes served by schools rated at least this (1 to 10, 0 = off), or in one school district, by its exact name on the site (Austin Independent School District). Applied by the site.
minSoldPrice, maxSoldPrice	For sold listings, the price they sold for, in US dollars, both bounds included. States that do not publish sale prices return nothing. Applied by the site.
commuteFrom, commuteMinutes, commuteMode	Only homes within 10 to 60 minutes of a point, by car or on foot. The point is written latitude,longitude (30.2672,-97.7431); the minutes and the way are ignored without it. Applied by the site.
floodRisk, fireRisk, windRisk, heatRisk, airRisk	Only homes whose risk is exactly the level you pick (minimal, moderate or severe), not "at most". Any = no filter. Applied by the site.
auctionOnly, shortSaleOnly, fractionalOwnershipOnly, matterportOnly, builderSpecHomesOnly, noPetsAllowed	Only auctions; only short sales; only homes sold in shares; only listings with a Matterport 3D scan; only builder spec homes (for sale); only rentals that take no pets (for rent, not together with petsAllowed). Applied by the site.
excludePropertyIds, excludePropertyTypes	Leave out listings by the id of a row (9954799904, up to 500) or whole property types (land, mobile). Not together with propertyType. Applied by the site.
extractDetails	Add price and tax history, schools, risk scores and value estimates (read 25 listings per extra request, charged as its own event). Off by default.
includeFloorplans, includeSchoolScores	Free, off by default, each needs extractDetails: the floor plans of the listing; the math and reading test results of each nearby school (a heavier answer, about 13 KB more per listing).
includeNearbyPlaces, nearbyPlaceCategories, nearbyPlacesRadiusMiles, nearbyPlacesPerCategory	Free, off by default, needs extractDetails: the shops and services around each listing, nearest first. Pick the categories (restaurants, grocery stores, coffee and tea by default; also cafes, bars, parks, gyms, fitness, health, pharmacies, banks, arts, beauty and spas, hotels, playgrounds, child care, convenience and hardware stores), the radius (1 to 25 miles, 2 by default) and how many places per category (1 to 10, 5 by default). The three settings do nothing without the box.
includePermitHistory, includeViewStats	Options, off by default: the building permits filed on the property with a roof-permit flag (charged per permit row), and the view / save / enquiry counts per period (charged per listing that has them). Either one turns extractDetails on.
countOnly	Free, off by default: only count how many listings each search has, without reading any. One number per search and one for the whole run, in the SEARCH_COUNTS record of the key-value store and in the log; the dataset stays empty and the run costs its start only. maxItems, extractDetails, the options, postedAfter, postedBefore, excludeKeywords, requirePhone and onlyNew are not used.
maxItems, maxItemsPerQuery	Stop after this many listings for the whole run (0 = unlimited); cap for EACH search (location × keyword, or search URL).
postedAfter, postedBefore	Listing date range: 2026-09-01, or a period before now (7 days, 2 weeks, 1 month, 24 hours).
excludeKeywords, requirePhone	Drop listings whose address, city or description contains one of these WHOLE words (case and accents ignored: pool drops "Pool and spa" but keeps "Liverpool Rd"); or whose agent has no phone number.
onlyNew, stateKey, resetState	Monitoring: only the listings never delivered under this memory key; resetState forgets the memory.
Advanced	proxyConfiguration (Apify proxy by default, included in the price; the residential proxy is not available), maxConcurrency, maxRequestsPerMinute, maxRequestRetries, debugLog.

📦 Output

One item, shortened to the fields a search result carries (the detail fields are cut here):

{
    "id": "7363437439",
    "listingId": "2816017",
    "url": "https://www.realtor.com/realestateandhomes-detail/7704-Fenton-Cv_Austin_TX_78736_M73634-37439",
    "status": "for_sale",
    "mlsStatus": "Active",
    "listPrice": 460000,
    "currency": "USD",
    "pricePerSqft": 285,
    "listDate": "2026-09-20T02:35:29.000Z",
    "propertyType": "single_family",
    "beds": 3,
    "baths": 2.5,
    "sqft": 1612,
    "lotSqft": 8015,
    "yearBuilt": 1982,
    "description": "Welcome to this stylishly updated home nestled on a private cul-de-sac lot...",
    "address": "7704 Fenton Cv",
    "city": "Austin",
    "stateCode": "TX",
    "postalCode": "78736",
    "countyName": "Travis",
    "latitude": 30.35,
    "longitude": -97.77,
    "agentName": "PATRICIA SMITH",
    "agentEmail": "patriciasmith@realtor.com",
    "agentPhone": "(512) 637-8277",
    "officeName": "Keller Williams Realty",
    "mlsName": "AUTX",
    "mlsNumber": "2816017",
    "photos": [
        "https://ap.rdcpix.com/.../1-m.jpg"
    ],
    "photoCount": 27,
    "daysOnMarket": 14,
    "searchUrl": "https://www.realtor.com/realestateandhomes-search/Austin_TX?page=1",
    "scrapedAt": "2026-09-20T12:00:00.000Z"
}

You can download the dataset in various formats such as JSON, HTML, CSV or Excel.

All 165 fields

Fields	Example
id, listingId, url	7363437439, 2816017, https://www.realtor.com/realestateandhomes-detail/...
status, mlsStatus, flags, tags	for_sale, Active, isNewListing, central_air
tagLabels	the tags the site shows with the listing, in clear text (Hardwood floors, View); tags keeps the codes
listPrice, currency, pricePerSqft, listPriceMin, listPriceMax	460000, USD, 285 — min / max for a new-construction price range
priceReducedAmount, priceReducedDate, lastSoldPrice, lastSoldDate	15000, 2026-09-12T15:04:11.207Z, 389900, 2020-08-24
listDate, lastUpdateDate, daysOnMarket	2026-09-20T02:35:29.000Z (what the date filters read), 14
pendingDate, lastStatusChangeDate	the day the listing went under contract and the last time its status changed (2026-10-30T20:18:54.000Z) — each when the site gives it
availableDate	rentals: the day the home can be moved into (2026-11-15)
moveInDate, applicationUrl, acceptsOnlineApplications	the day a new home can be moved into (2027-04-30), the link to apply online for a rental, and true when the rental takes online applications — each when the site gives it
applicationFee, securityDeposit, petDeposit, petFee, petRent, leaseTermMonths	detail — rentals: the application fee, the security deposit, the pet deposit, the one-time pet fee and the monthly pet rent in US dollars (55, 500, 300, 200, 35), and the lease lengths offered in months ([6, 12]) — each when the site lists it
propertyType, propertySubType, yearBuilt, stories, garage	single_family, condo, 1982, 2, 2
beds, baths, bathsFull, bathsHalf, sqft, lotSqft, hoaFee	3, 2.5, 2, 1, 1612, 8015, 250
description, features	the agent's own text; feature lines grouped by category (Appliances, Interior…)
address, city, state, stateCode, postalCode	7704 Fenton Cv, Austin, Texas, TX, 78736
streetNumber, streetName, streetSuffix, streetDirection, unit	the address cut in parts (7402, Rook, St, W, Unit 21) — each when the site gives it; address stays the whole line
countyName, countyFips, neighborhoods, latitude, longitude, streetViewUrl	Travis, 48453, Oak Hill, 30.35, -97.77
agentName, agentEmail, agentPhone, agentPhones, agentPhotoUrl, agentWebsite	PATRICIA SMITH, patriciasmith@realtor.com, (512) 637-8277, every number with its type, the agent's photo, the agent's website as a link. With detail, an empty e-mail or phone is filled with the one the MLS gives for the seller's agent, only when that agent has the same name (or the listing names no agent, then the name comes from the MLS too); otherwise the MLS contact stays in mlsAgents
officeName, officeEmail, officePhone, officeWebsite, brokerName, mlsName, mlsNumber, listingSource	Keller Williams Realty, AUTX, 2816017, mls (or a builder, new_home, or a rental company, community / unit_rental)
agentType, agentNrdsId	seller (or co_seller, community, management), the agent's NRDS member number (785030693), when the site gives it
agentLicense, agentProfileUrl	detail — the agent's state license number (0664627) and their realtor.com page — each when the site gives it
buyerAgentName, buyerOfficeName, coAgentName, mlsAgents	for a sold home, the buyer's agent and their office; the second agent of the seller; and every agent the MLS lists, each with a name, a role (seller, co_seller, buyer, co_buyer) and an office. Some MLSs give the office only, so a name can be empty. With detail, the buyer's agent also comes with the office as the site writes it (Jane Doe, Example Realty), and each agent of mlsAgents also has an e-mail, a phone, the MLS's agent id, the office id and the office phone, when the MLS gives them
buyerAgentEmail, buyerAgentPhone, buyerAgentLicense, buyerAgentWebsite, buyerOfficeEmail, buyerBrokerName	detail, sold homes — the buyer's agent: e-mail, phone (the one the site marks primary), state license, website, the e-mail of their office and their broker — each when the site gives it
daysOnMls, mlsBoard, providerUrl, createdAt, timezone	detail — the days the MLS counts since the listing was entered, the board of realtors that runs the MLS (Austin Board Of Realtors), the listing on the website of the agency or the MLS, the day the site created the listing and the property's time zone (America/Chicago) — each when the site gives it
virtualTours, matterportUrl, rentSpecials	detail — every virtual tour (link, kind and category), the first Matterport link, and the specials of a rental (Enjoy up to 8 Weeks Free, with the start and end day when given); empty when there are none
mlsFullName, listingKey, contractDate, indexedAt	the full name of the MLS (UnlockMLS), the key the MLS gives the listing, the day the sale went under contract (2026-11-11) and the time the site indexed the listing — each when the site gives it
builderName	the home builder of a new-construction listing (PERRY HOMES)
community	the builder's program a new-construction listing belongs to, as one object with the same keys every time: name, page, text, builder and its site, state, number of homes, price / beds / baths / garage / size ranges, driving directions, amenities and the program's promotions; empty for any other listing
promotions, promotionLabel	the promotions of the listing itself (title, text, link), and the short label the site puts on the search card ($8K closing); empty when there is none
mlsDisclaimer	the notice the listing's MLS asks to show with it (© 2026, Unlock MLS, All Rights Reserved.), when the site gives it
buildingName, subdivision, bathsThreeQuarter	the name of the building or rental community (461 Dean), the subdivision (Heritage Oaks Sec 2-B, or the county's name for it) and the three-quarter baths (1) — each when the site gives it
yearRenovated, buildingUnits, zoning	detail — the year of the last renovation (2017), the number of units of the building (122) and the zoning code (A-2) — each when the site gives it
brandingName, brandingType, brandingLogoUrl, brandingAccentColor	the first brand the site shows on the listing: its name, its kind (Agent, Office or Builder), its logo and its accent color
photos, photoCount, virtualTourUrl, openHouses, petPolicy	27 photo URLs, a Matterport link, the scheduled open houses, which pets a rental takes
hasMatterport	true when the listing has a 3D tour, false when the site says it has none
photoTags, photoDetails, primaryPhotoDetails	the labels the site puts on the photos (kitchen, bathroom, yard), and for each photo, and for the main one, its link, its type and its labels with a probability (0.999)
otherListings	the other listings of the same home (earlier sales, rentals, relistings), each with its id, status, price, dates and whether it is the main one
publicRecord	the county's public record of the home (the tax roll) as one object with the same keys every time (a key is empty when the county has no value; the whole object is empty when the site has no record): record id and update day, subdivision, census tract, property type, beds, baths, sqft, lot size, stories, year built, garage. With detail it also has the parcel number, acres, building / gross / garage / first-floor / basement sqft, lot width and depth, rooms, units, year renovated, heating, cooling, electrical, water source, fireplace, construction, exterior, roofing, pool, zoning, view, styles and the legal description
rentalUnits, unitCount	rental communities: one entry per floor plan (name, beds, baths, sqft, rent, available day, first photo) and the number of plans. With detail each plan also has its top rent and deposit, and its units (number, floor, rent, beds, baths, sqft, available day, vacancy, amenities); a very large community is cut at 60 plans and 40 units per plan. Empty for any other listing. For a building, unitCount can also come from the detail
priceHistory, taxHistory, estimates, estimatesHistory, estimatesForecast	detail — 39 listing events, 17 years of tax (each with the assessed and the market value), 3 value estimates with past and forecast
schools, floodScore, floodSeverity, femaZone, fireScore, noiseScore, noiseCategories	detail — nearby schools with their rating, the parents' rating and review count, the school's page and whether the home is assigned to it, students per teacher, each school's grades, address, position, GreatSchools id, NCES code and phone, and with includeSchoolScores the math and reading share at the state standard and its year; flood score and level (minimal), FEMA zone, fire / noise scores
schoolDistricts	detail — the school districts the home is assigned to: id, name, NCES code, GreatSchools id, number of schools and students, the district office address and its county
floorPlans	detail with includeFloorplans — the floor plans of the listing, each with its kind (2d, 3d, interactive, cubicasa), type, link and thumbnail; an empty list when it has none
nearbyPlaces	detail with includeNearbyPlaces — the shops and services around the listing, nearest first: name, categories, rating (null when nobody rated it), number of reviews, distance in miles, phone, address, position, website and link to its page; an empty list when not asked or when nothing is around
neighborhoodStatsName, neighborhoodMedianDaysOnMarket, medianRentPrice	detail — the first neighborhood of the listing with market figures (Southeast Austin), its median days on market and its median monthly rent in USD
monthlyPayment	detail — mortgage estimate
mortgageRate, mortgageLoanAmount	detail — the interest rate in percent (7.01) and the loan amount in US dollars (889000) behind that estimate
mortgageRateThirtyYearFixed, mortgageRateFifteenYearFixed, mortgageRateFiveYearArm	detail — the site's average rate of the day, in percent, for a 30-year fixed, a 15-year fixed and a 5/1 adjustable loan (7.01, 6.146, 6.45)
rentEstimate, rentEstimateLow, rentEstimateHigh	detail — the monthly rent the site estimates for the home, with its range (2502, 2279.25, 2724.75)
stats	option includeViewStats — views, clicks, saves, shares and enquiries per period
buildingPermits, hasRoofPermit	option includePermitHistory — each permit (project, type of work, date, status); true when one is a roof job
searchRank, searchPage, totalResults	the place of the listing in its search (1 = first result of that search), the results page it was on, and the number of listings the site announced for the search — empty for a listing URL given in startUrls
searchUrl, scrapedAt	the search it was found through, ISO timestamp

💡 Tips

How to get more results

One search returns at most 10,000 listings — that is the site's own limit, not the Actor's. Above it, the Actor splits the search into price bands, and each band again until every piece fits — sold homes into sale-date ranges instead, since many of them have no price — so a whole state comes out in one run. The few listings with no price at all (about 1 in 1,000 rentals, fewer for sale) cannot be reached by a price band; sold homes lose none. The count the site shows also counts twice a home listed by two sources (two MLS, or an MLS and a rental manager): it comes out once, under one id and one URL. Measured on whole searches: Harris County for sale, 29,279 listings for a count of 29,403; Arizona rentals, 35,456 for 35,842. You can also split a search yourself by putting several ZIP codes in locations. Set maxItems to 0 to take everything a search has.

How to reduce costs

The price is per listing, so the levers are maxItems, maxItemsPerQuery, the filters (a filtered-out listing costs only its filter check, $0.04 per 1,000, $0.01 on the Gold plan) and onlyNew for recurring runs (you never pay twice for the same listing). Leaving extractDetails off keeps listings at the lower price: the search results already carry description, photos and the agent's contact details.

Count the listings of a search first

Tick Count only (countOnly) to see how many listings each of your searches has before you read them. The run answers in seconds, saves nothing to the dataset, and charges only its start. Open the SEARCH_COUNTS record of the run's key-value store: one entry per search with its place, keyword, status, link and total, and totalListings for the whole run (a listing found by two searches counts twice there). A search the site could not count has total: null and a note (never 0). A search with a total above 10,000 is counted whole, even though a normal run reads it in pieces. A listing URL in startUrls has no count. The filters you set (place, status, price, beds, map area…) are part of the search and are counted; the post-filters (postedAfter, postedBefore, excludeKeywords, requirePhone) are not.

Several searches in one run

Fill locations and / or searchQueries: the Actor runs one search per location × keyword (3 keywords × 4 cities = 12 searches, up to 500 per run). The single location and query fields still work and are added to the lists. A listing found by several searches is saved — and charged — once. Set maxItemsPerQuery to give every search its own cap: without it the first searches can use up the whole maxItems budget. You can also paste several search URLs into startUrls: each one is a search of its own, with the same cap.

Monitoring: only the new listings

Tick Only new listings (onlyNew) and schedule the Actor. The first run returns everything; each later run skips the listings already delivered: they are not saved, not charged, and no detail request is made for them. The memory lives in a named key-value store of your account (realtor-listings-scraper-seen, up to 150,000 listings per key) and is only updated with listings that really reached the dataset, so a failed run never hides anything. Give each schedule its own stateKey (two schedules sharing a key would hide each other's listings), and tick resetState once to start over. The searches of the form are then sorted by listing date, newest first (rentals: in the order they were published, which also reaches the rentals that carry no date; a pasted URL keeps its own sort), and a search stops once it meets 200 listings in a row that you already have — or, under a lower maxItems / maxItemsPerQuery, 80 % of that cap in all (a listing of the block the last run delivered may have been sold or relisted since). Realtor.com dates a listing by the day it first went on the market and never moves that date afterwards, so a listing you already have cannot climb back above a new one. Rental buildings (apartment communities) are the exception: the site puts them back on top each time they are updated, so they never count in those 200, and one you already have is still skipped.

Filter by listing date

postedAfter and postedBefore take a date (2026-09-01, the whole day is included, site time zone) or a period before now (7 days, 2 weeks, 1 month; via the API also 24 hours or a full ISO date-time). The filter reads listDate, the moment the listing went on the market; a listing without one is dropped as soon as a date bound is set. Filtered-out listings are not saved and do not count in maxItems (each listing checked costs the filter fee, see pricing); the run summary tells how many were filtered. With postedAfter, the searches of the form are sorted by newest and stop at the first page that is entirely too old; a URL pasted into startUrls keeps its own sort, and stops early only when it is sorted by newest listing date — otherwise it is read to its end, each listing checked.

🔌 Integrations and API

Call the Actor via the Apify API, the JavaScript or Python clients, or connect it with integrations and webhooks (Make, Zapier, n8n, Google Sheets, Slack, Airtable…). The dataset can be fetched as JSON or CSV from any tool.

🤖 Use with AI agents (MCP)

AI agents (Claude, ChatGPT, Cursor…) can find and run this Actor through the Apify MCP server, billed to their Apify account like any run. It returns one item per Realtor.com property listing. Actor id: nice_dev/realtor-listings-scraper; MCP server with this Actor only: https://mcp.apify.com/?tools=fetch-actor-details,nice_dev/realtor-listings-scraper.

Smallest input, for a cheap first call:

{
    "location": "Austin, TX",
    "maxItems": 10
}

Key output fields: url, status, listPrice, beds, baths, sqft, address, agentPhone.

Cost: $0.45 per 1,000 listings plus $0.01 per run start ($0.42 per 1,000 on the Gold plan); Extract details, the permit and view-statistics options and the Actor's own filters cost extra, see the pricing section above. Cap each call with maxItems and, through the API, with the run option maxTotalChargeUsd.

❓ FAQ

Is it legal to scrape Realtor.com?

The Actor only reads what Realtor.com shows publicly to any anonymous visitor. It logs in to nothing. Results contain the contact details of licensed real-estate professionals acting in their trade, and personal data is protected by GDPR and by US state privacy laws: do not store it without a legitimate reason, and do not use it for unsolicited marketing where that is restricted. Listing content comes from local MLS boards and its redistribution can be limited by their rules. You are responsible for using the data in compliance with Realtor.com's Terms of Use and applicable law. This Actor is not affiliated with Realtor.com, Move, Inc. or the National Association of REALTORS®.

Does it need a login or a proxy?

No login. The proxy is included in the price: leave the default setting (the residential proxy is not available). A request the site turns away is retried at once on a new proxy session (without a proxy, after a pause of 5 seconds, doubled at each retry up to 150 seconds).

Which fields can be empty?

Realtor.com does not publish everything for every listing, and the Actor never invents a value: a field it has no data for is null or an empty list. Measured on 800 real listings of every kind: agentEmail and agentPhone are there on about 6 listings in 10 (almost always on sold listings, less often on new ones), beds / baths / sqft are missing on land and mobile homes, and lastSoldPrice is null in non-disclosure states such as Texas and Utah, where sale prices are not public records. With extractDetails on and measured on 30 listings: price history and schools on all of them, tax history on 20, value estimates on 22. With the options on (26/09, 450 listings): building permits on 169 of 300 listings in Austin (5 rows on average, up to 20) and 116 of 150 in Tampa, where 69 carry a roof permit; view counts on 426 of 450 — a brand-new listing has none yet, the site starts counting a few days after it goes up.

Is the data safe to open in Excel or to show on a web page?

Descriptions are the agents' own words, copied as they are, and a text can begin with -, +, = or @ (a phone number, a line such as -20% price cut): Excel and Google Sheets may read such a cell of a CSV file as a formula or as a number. The Actor leaves the text as it is, so that the JSON and the API give the real value: when you open a CSV, import these columns as text. Every URL field (url, photos, virtualTourUrl, streetViewUrl, agentPhotoUrl) holds an http(s) URL or null — nothing else gets through. On a web page, escape every field like any text written by a stranger.

Known limitations

• Site filters that are not in the input can still be used by pasting a filtered search URL into startUrls. A URL whose filters the Actor cannot apply is refused by name instead of being run without them.
• One search returns at most 10,000 listings on the site itself; bigger searches are split automatically, and maxItemsPerQuery still counts for the whole search, not for each piece. Past 10,000, the listings come band by band: the sort order holds within each band, not across the whole search.
• daysOnMarket is counted from listDate: Realtor.com leaves its own field empty. For a sold home it stops at the sale day (lastSoldDate).
• onlyNew remembers listing ids, not their content: a listing whose price changed is not returned again.
• Rentals: part of the rental listings carry no listing date on Realtor.com (12 % in Austin, 61 % across Arizona), so their listDate and daysOnMarket are empty. onlyNew still returns the new ones (rentals are read in the order they were published); postedAfter / postedBefore leave them out.
• Sold homes: many sales were never listed and come from public records (30 % of Texas sales, a third to a half in some cities). They have no listPrice, listDate, MLS number, agent or description (listingSource is empty); sorted by listing date, sale date or price they come last, sorted by size or last update they are mixed in.
• "Sold in the last N days" counts from the day the run started (New York time), for its whole length — a resurrected run keeps that day.
• Off market: an expired listing may carry no listing date (its listDate and daysOnMarket are then empty, and postedAfter / postedBefore leave it out); flags.isForRent tells an expired rental from an expired sale. Past 10,000 listings the search is cut by price, which leaves out the few listings that have no price.
• Ready to build: a builder plan has no listingId (the id is the property id) and flags.isPlan is true; some plans have no price. mlsOnly and builderSpecHomesOnly cannot be used with it.
• Off market and ready to build with onlyNew: the run reads the whole search every time and skips what it already delivered (these listings are not ordered by the moment they appear), so it takes as long as a full run without onlyNew.
• Two runs sharing the same stateKey at the same time may both return the same new listing.

A run that reaches its timeout stops itself about 45 seconds before it: no new page is asked, what it read is saved and, with onlyNew, remembered, and the run ends Succeeded with "Stopped before the run's timeout". Resurrect it to go on from there, or give the next run a longer timeout (Run options).

A run the platform stops without warning (out of memory)

• Resurrect it: it goes on from where it stood at most a minute before the stop. What it had read since is read again, and the listings already saved are skipped: none is delivered or charged twice, and maxItems still counts them.
• With onlyNew, the memory is saved after each write (at most once every 2 seconds): resurrect the stopped run and the listings it had saved meanwhile join the memory; leave it stopped for good, and the next run may return the listings of its last 2 seconds once more.

Something doesn't work?

The last line of the log counts the listings saved, filtered out and no longer on Realtor.com (removed while the run was reading them), and the requests that failed after every retry. Those requests and the removed listings are listed, with the reason, in the FAILED_REQUESTS record of the run's key-value store. A run that saved nothing and had failed requests fails, and its last message gives the cause (a location the site does not know says so, instead of "run it again"). A run that saved some listings fails too when at least as many requests failed for good as were read: a green run with a short dataset would hide an outage. One failed request among many is only a warning.

If Realtor.com changes its pages, you are told instead of paying for blank rows. A results page that counts listings but gives none the Actor can read is an error (listed in FAILED_REQUESTS), never a quiet "No listings found". If the first 20 listings read all lack their URL, status, price or listing date (rentals aside), last update date, property type, address, city, state, ZIP code, county, MLS number or — with details — price history, the run saves nothing more, stops and fails, and its last message names the missing field: at most those first listings are charged. A listing that postedAfter / postedBefore drops because it has no date at all counts among those 20. With details, if the first 20 listings of the run all come back as no longer on Realtor.com, the run stops and fails the same way, with nothing charged.

🛟 Support

Open an issue in the Issues tab with a link to your run: the run log and the FAILED_REQUESTS record of the key-value store show exactly which URLs failed and why.