Zameen Scraper [From $0.9💰] | Listings | Agencies | PK Data

Zameen Scraper — Pakistan Property Listings & Agent Data From zameen.com

Extract ~237,490 active Zameen.com listings and ~15,423 agencies into structured JSON — price in PKR, area in marla / kanal / sq ft, beds & baths, society / block, photos, full descriptions and agent mobile / landline / WhatsApp numbers — by querying Zameen's own Elasticsearch search backend directly. Built for property price monitoring, Pakistan property databases and real-estate lead generation — no browser, no anti-bot bypass.

---

Main Use Cases: Property Price Monitoring, Pakistan Property Database & Real Estate Lead Generation

• Property price monitoring — pull the freshest sale and rent asking prices (sort: newest, purpose, priceMin / priceMax) city by city (Lahore, Karachi, Islamabad, …), then diff runs on listingId + pricePkr to catch price drops.
• A Pakistan property database — sweep the full ~237,490-listing index with search_after pagination (no 10,000-result ceiling) and export JSON, CSV or Excel from the Apify dataset.
• Real estate market research — filter by purpose, category, rooms, price, area and createdAt; city, society, block and locationPath[] structure rows for city- and society-level analysis (DHA Lahore plot rates, Karachi commercial rates, Islamabad rents).
• Real estate lead generation — mode: agencies returns up to 15,423 agencies with mobile, landline and WhatsApp numbers, listing counts, licenses and tier.
• Rental yield analysis — sale and rent rows share one schema (purpose, rentFrequency, pricePkr, areaMarla) — gross-yield tables are a join away.
• Overseas Pakistani property investment research — PKR prices, marla/kanal areas and coordinates for PK/Gulf investor analysts, no browser needed.

---

How The Zameen Scraper Works (Elasticsearch Backend, No Browser)

Zameen is a Next.js-style server-rendered app: any plain HTTP GET returns the full application state inline as window.state. This Actor calls the search cluster it describes — Zameen's own Elasticsearch _search API — over plain HTTP with size: 1000 pages and search_after pagination, maps each _source document to one dataset row, and optionally (fetchFullDetails: true) runs a CheerioCrawler pass over the property pages' window.state blobs to add the full description, every photo URL and the complete phone set. No Playwright/Puppeteer, no Cloudflare challenge, no paid anti-bot API. Rows are recordType: "property" or recordType: "agency".

---

Input Parameters

All fields are optional (no required input); the defaults below run immediately.

Field	Type	Required	Default	Notes
mode	search | detail | agencies	No	search	search = listings, detail parses propertyUrls, agencies = agent/agency directory.
purpose	for-sale | for-rent	No	— (UI prefill for-sale)	Empty = both sale and rent.
city	string	No	—	Exact Zameen location name: Lahore, Karachi, Islamabad, Rawalpindi, Multan, Faisalabad, Peshawar, Gujranwala, …
category	string	No	—	Property category, e.g. Homes, Plots, Offices, Commercial, Agricultural, Industrial.
bedroomsMin	integer	No	—	Minimum bedrooms (0–20).
bedroomsMax	integer	No	—	Maximum bedrooms (0–20).
bathsMin	integer	No	—	Minimum bathrooms (0–20).
bathsMax	integer	No	—	Maximum bathrooms (0–20).
priceMin	number	No	—	Absolute PKR, e.g. 20000000 = 2 crore (min 0).
priceMax	number	No	—	Absolute PKR, e.g. 50000000 = 5 crore (min 0).
areaMinSqft	number	No	—	Square feet; converted to m² internally (1 marla = 225 sq ft, 1 kanal = 4,500 sq ft).
areaMaxSqft	number	No	—	Square feet; converted to m² internally (same factors).
keyword	string	No	—	Free-text match against title, short description, keywords and location path (e.g. DHA).
sort	newest | oldest | price-asc | price-desc | area-desc	No	newest	newest = freshest inventory, best for lead-gen and price tracking.
maxListings	integer	No	100	Hard cap for mode: search (1–20,000); search_after scales past the 10k window. Free plans cap 10 rows.
maxAgencies	integer	No	25	Hard cap for mode: agencies (1–5,000; Zameen publishes ~15,400 agencies).
maxDetailPages	integer	No	20	Property pages getting the Cheerio full-detail pass (1–500). Free plans cap 10.
fetchFullDetails	boolean	No	false	Enrich each listing with full description, all photo URLs, video/panorama counts and contact numbers — still no browser (window.state blob).
propertyUrls	string[]	No	[]	Required only for mode: "detail" — explicit https://www.zameen.com/Property/...html URLs.
proxy	object	No	{ "useApifyProxy": false }	Standard Apify proxy picker — choose group/country in the UI. Not needed by default: the backend is unauthenticated-by-WAF.

---

Output Fields

The shipped dataset schema declares an empty fields map and one table view — Overview — with columns recordType (Type), title (Title), name (Agency), city (City), purposeLabel (For), priceFormatted (Price), areaMarla (Area), bedrooms (Beds), bathrooms (Baths), url (Listing URL) and profileUrl (Agency URL). Every row carries recordType: "property" or "agency".

Property row — 58 fields, grouped:

• Identity — listingId, externalId, dldPropertySK, referenceNumber, url, slug, title, titleUrdu
• Deal — purpose ("for-sale" | "for-rent"), purposeLabel, pricePkr, priceFormatted, rentFrequency, category, subCategory
• Area — areaSqm, areaSqft, areaMarla, areaKanal
• Rooms — bedrooms, bathrooms
• Location — country, province, city, society, block, locationName, locationHierarchy, locationPath[], geoLat, geoLng
• Status — completionStatus, furnishingStatus, occupancyStatus, ownershipStatus
• Media & tags — amenities[], keywords[], photoCount, coverPhotoUrl, photos[{url,id,title,order}], videoCount, panoramaCount
• Text — shortDescription, description, descriptionHtml
• Contact — contact { contactName, mobile, phone, whatsapp, allPhoneNumbers[], allMobileNumbers[] }
• Listing agency — agency { agencyId, externalId, name, slug, profileUrl, logoUrl, type, product, tier, tr }
• Flags & meta — isVerified, verificationStatus, product, state, createdAt, updatedAt, reactivatedAt, scrapedAt

Agency row — agencyId, name, slug, profileUrl, logoUrl, city, cities[], geoLat/geoLng, licenses[], agentsCount, listingsTotal / listingsForSale / listingsForRent, contact { mobile, phone, allPhoneNumbers[], allMobileNumbers[] }.

Area units (marla, kanal, sq ft)

Zameen stores area in square metres (verified against live documents):

Document	area (raw)	Correct
5 marla house	104.51592	1,125 sq ft
20 marla plot	418.06368	4,500 sq ft = 1 kanal
8 kanal commercial	3344.50944	36,000 sq ft

It emits areaSqft, areaMarla (1 marla = 225 sq ft) and areaKanal (1 kanal = 20 marla) alongside areaSqm.

---

Example Input And Output

Freshest Lahore sale listings with full descriptions:

{
  "mode": "search",
  "purpose": "for-sale",
  "city": "Lahore",
  "sort": "newest",
  "maxListings": 500,
  "fetchFullDetails": true,
  "maxDetailPages": 50
}

3–4 bed homes between 2 and 5 crore in Karachi:

{ "mode": "search", "purpose": "for-sale", "city": "Karachi", "category": "Homes",
  "bedroomsMin": 3, "bedroomsMax": 4, "priceMin": 20000000, "priceMax": 50000000, "maxListings": 200 }

Agency lead list:

{ "mode": "agencies", "city": "Lahore", "maxAgencies": 500 }

One property row:

{
  "recordType": "property",
  "listingId": 2961458,
  "url": "https://www.zameen.com/Property/Islamabad-G-13-5-Marla-House-for-sale-2961458.html",
  "title": "5 Marla House for sale in G-13, Islamabad",
  "purpose": "for-sale",
  "purposeLabel": "For Sale",
  "pricePkr": 27000000,
  "priceFormatted": "PKR 2.7 Crore",
  "category": "Homes",
  "areaSqm": 104.51592,
  "areaSqft": 1125,
  "areaMarla": 5,
  "areaKanal": 0.25,
  "bedrooms": 3,
  "bathrooms": 4,
  "city": "Islamabad",
  "society": "G-13",
  "locationName": "G-13, Islamabad",
  "geoLat": 33.668,
  "geoLng": 72.986,
  "photoCount": 12,
  "shortDescription": "Well-located 5 marla house in G-13 near amenities.",
  "contact": {
    "contactName": "Example Realtor",
    "mobile": "+92 300 1234567",
    "whatsapp": "+92 300 1234567",
    "allPhoneNumbers": ["+92 300 1234567"]
  },
  "agency": {
    "agencyId": 48215,
    "name": "Example Estates",
    "profileUrl": "https://www.zameen.com/agents/example-estates/"
  },
  "isVerified": true,
  "scrapedAt": "2026-10-07T09:15:00.000Z"
}

---

Zameen API Alternative — Structured Property Listings Data From zameen.com

There is no official Zameen developer API. A plain curl of any page returns the application state inline as window.state, carrying the connection details of the search cluster Zameen's own frontend queries:

$curl -s https://www.zameen.com/ | grep -o 'window.state = .\{0,900\}'

The state key is named algolia (historical) but activeBackendName is "elastic" — the live backend is a self-hosted Elasticsearch cluster on AWS, not Algolia; the dns-prefetched Algolia DSN serves type-ahead only.

Exact endpoints:

Purpose	Index	Endpoint
Property listings	zameen-production-ads-en	POST https://search-zameen-live-kiqsqbqxiuzpzkqbjwkboe32g4.eu-west-1.es.amazonaws.com/zameen-production-ads-en/_search
Agencies / agents	zameen-production-agencies-en	POST https://search-zameen-live-kiqsqbqxiuzpzkqbjwkboe32g4.eu-west-1.es.amazonaws.com/zameen-production-agencies-en/_search

Auth: HTTP Basic with the credentials Zameen publishes in window.state — over plain-HTTP JSON _search, not behind Cloudflare, the web WAF, cookies, tokens or any browser challenge.

Why this beats rendering the site:

	Site HTML (browser)	This Actor's backend
Requests per 1,000 listings	~1,000 page loads	1 _search (size: 1000)
Anti-bot	Cloudflare + possible challenge	none
Memory / compute	full Chrome (~1 GB+)	plain HTTP client (256 MB)
Enrichment	client-side fetch	full record already in _source

Verified cluster capabilities: size up to 5000 (Actor uses 1000 per page); search_after deep pagination → no 10,000 result_window ceiling, so a full 237k sweep is possible; aggregations for facet discovery (e.g. location.name gives every city / society); filterable fields purpose, state, location.name, category.name, rooms, baths, price, area, createdAt; sortable fields createdAt, price, area, id (listings) and stats.adsCount, id (agencies).

Field-mapping gotchas already handled in code: location.name / category.name are the mapped paths (there is no .keyword sub-field — a .keyword term query silently returns 0 hits), and the agency sort field is stats.adsCount (no root-level adsCount).

Secondary source — CheerioCrawler, still no browser. The listings index only carries a truncated shortDescription; the full text and every photo URL live in the property page's window.state blob (state.property.data = description, photos[], floorPlans, panoramas, videos, primaryPhoneNumber, …). With fetchFullDetails: true the Actor runs a CheerioCrawler (plain HTTP + JSON parse, no DOM scraping) over the first maxDetailPages listings — property pages return HTTP 200 to a plain got-scraping request, so no browser is needed.

Public URL shapes:

• Listing detail: https://www.zameen.com/Property/{slug}.html (field: link; carried per row as url + slug)
• Agency profile: https://www.zameen.com/agents/{slug}/ (verified live: /agents/<slug>/ → 200, /agency/<slug> → 404)

Maintainability. If Zameen rotates the cluster or credentials, update two constants in src/routes.ts (ES_URL, ES_CREDENTIALS) plus the index names in the same file, then re-discover them with the window.state curl above. If the blob ever moves, the Actor degrades gracefully: search-cluster rows still return (short description + cover photo) without the enrichment pass.

---

Run Zameen Scraper From The Apify API, Console Or Schedule

Start runs from the Console or through the API:

curl -X POST "https://api.apify.com/v2/acts/6HsHbEjMz6yeJ1pRn/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"mode":"search","city":"Lahore","sort":"newest","maxListings":200}'

Then poll GET /v2/actor-runs/<runId> and read rows from GET /v2/datasets/<datasetId>/items?format=json. The dataset UI exports JSON, CSV, Excel; Google Sheets via Zapier / Make.

Cost & limits:

• Memory: 256 MB (max 512 MB) — plain HTTP, no browser.
• Free vs paid: free accounts are capped at 10 rows per dataset via APIFY_USER_IS_PAYING; paying accounts run the full configured caps.
• Runaway protection: hard caps in code (HARD_MAX_LISTINGS, HARD_MAX_AGENCIES, HARD_MAX_DETAIL_PAGES), maxRequestsPerCrawl, maxRequestRetries: 2, timeouts, and terminating search_after pagination.
• No paid anti-bot API required — the Elasticsearch backend is unauthenticated-by-WAF.

Pay-per-event billing:

Event	Price
Listing	$0.0009 per property row
Agency	$0.0009 per agency row
Actor Start	$0.00005 per run

---

Integrations & Automation

• Apify API — start, poll and read results programmatically.
• Webhooks — fire on actor.succeeded / actor.failed to push rows into a warehouse, Google Sheets or Slack.
• Zapier / Make — turn a new dataset item into a CRM record or deal-alert email without code.
• Schedules — run daily for price monitoring (e.g. cron 0 6 * * *) or weekly for a market digest; keep maxListings stable so dataset diffs stay meaningful.

---

Related Actors

• zameen-scraper
• zameen-scraper
• zameen-property-scraper
• zameen-real-estate-scraper
• zameen-com-scraper

---

FAQ

Does Zameen.com have an official API?

No — a public API-marketplace FAQ asks that verbatim and answers "no official developer API". This Actor queries the site's own Elasticsearch _search backend — the endpoint the Zameen frontend consumes — returning the same records as dataset rows.

Why use this actor instead of the official API?

There is no official API. Versus rendering the site yourself: 1 _search request (size: 1000) covers what costs ~1,000 page loads, memory stays at 256 MB instead of ~1 GB+ of Chrome, no Cloudflare challenge is ever hit, the full record already sits in _source, and search_after removes the 10,000-result window so a full ~237,490-listing sweep is possible.

How do I check property prices in Lahore, Karachi and Islamabad?

Set city to the exact Zameen name (Lahore, Karachi, Islamabad, Rawalpindi, Multan, Faisalabad, Peshawar, Gujranwala) and sort with price-asc / price-desc, or newest for current inventory. Rows carry pricePkr, priceFormatted, areaMarla / areaKanal / areaSqft and society / block for city-level price tables.

How do I get property prices in DHA Lahore?

Run mode: "search" with city: "Lahore" and keyword: "DHA" (the keyword matches the location path, so DHA phases surface), then group rows by society / block or filter locationPath[].

How can I export Zameen listings to CSV, JSON or Google Sheets?

Run mode: "search", then export the Apify dataset as JSON, CSV or Excel, or push rows to Google Sheets via Zapier / Make.

How do I monitor new listings and price drops?

Schedule daily or weekly (sort: "newest", stable maxListings), diff datasets on listingId + pricePkr + updatedAt, and use webhooks to alert on completion.

Is Zameen.com data reliable for property valuations?

Treat it as asking-price evidence, not transaction prices — analysts warn portal figures are asking-price proxies, and Pakistan has no central open transaction registry, so listing-portal aggregates are the de-facto source.

What is the rental yield in Lahore, Karachi or Islamabad?

Run both purposes (purpose empty = sale + rent), then compute gross yield as annualised rent ÷ sale price from pricePkr, rentFrequency and areaMarla.

How do I get real estate agent phone numbers in Pakistan?

mode: "agencies" returns up to ~15,423 agencies with contact.mobile, contact.phone, whatsapp, licenses, listing counts and tier, plus city / cities[]. Business contact details published for marketing — B2B lead generation, not unsolicited consumer outreach.

What are the alternatives to Zameen Scraper?

Portals: Graana, OLX Pakistan's property vertical and eproperty.pk. Store actors: haketa/zameen-scraper, solidcode/zameen-scraper, logiover/zameen-property-scraper, crawlerbros/zameen-real-estate-scraper, shahidirfan/zameen-com-scraper.

---

SEO Keywords

zameen scraper, zameen api, zameen property listings data, zameen data export, zameen alternative, pakistan property database, real estate data extraction, property price monitoring, real estate market research, rental yield analysis, real estate lead generation, real estate agent contact list pakistan, property price tracking, pakistan real estate market, pakistan property portal, lahore property prices, karachi property prices, islamabad property prices, property prices in dha lahore, plot for sale, commercial property rates karachi, rent prices islamabad, overseas pakistani property investment

For AI Agents & LLM Apps

Purpose: query Zameen.com's own Elasticsearch backend (optionally its public property pages) → one JSON row per property listing (58 fields) or agency: PKR prices, marla/kanal/sq ft areas, contact details — no browser.

Minimal input:

{ "mode": "search", "city": "Lahore", "maxListings": 25 }

Variant — agency directory mode:

{ "mode": "agencies", "city": "Karachi", "maxAgencies": 100 }

Output fields: recordType; property rows — listingId, externalId, dldPropertySK, referenceNumber, url, slug, title, titleUrdu, purpose, purposeLabel, pricePkr, priceFormatted, rentFrequency, category, subCategory, areaSqm, areaSqft, areaMarla, areaKanal, bedrooms, bathrooms, country, province, city, society, block, locationName, locationHierarchy, locationPath, geoLat, geoLng, completionStatus, furnishingStatus, occupancyStatus, ownershipStatus, amenities, keywords, photoCount, coverPhotoUrl, photos, videoCount, panoramaCount, shortDescription, description, descriptionHtml, contact{contactName, mobile, phone, whatsapp, allPhoneNumbers, allMobileNumbers}, agency{agencyId, externalId, name, slug, profileUrl, logoUrl, type, product, tier, tr}, isVerified, verificationStatus, product, state, createdAt, updatedAt, reactivatedAt, scrapedAt; agency rows — agencyId, name, slug, profileUrl, logoUrl, city, cities, geoLat, geoLng, licenses, agentsCount, listingsTotal, listingsForSale, listingsForRent, contact{...}.

Behaviors an agent should know:

• mode defaults to search; mode: "detail" requires propertyUrls; "agencies" switches rows to agency records (billed as Agency).
• Empty purpose means both sale and rent; sort defaults to newest.
• Caps: maxListings 100 (max 20,000), maxAgencies 25 (max 5,000), maxDetailPages 20 (max 500). Free plans return only 10 rows regardless of input (APIFY_USER_IS_PAYING).
• fetchFullDetails defaults to false; true adds one plain-HTTP request per enriched listing, capped by maxDetailPages — still no browser.
• priceMin/priceMax are absolute PKR; areaMinSqft/areaMaxSqft convert to m² internally (1 marla = 225 sq ft, 1 kanal = 4,500 sq ft).
• city must match Zameen's exact spelling; this backend has no .keyword sub-field — filter via the Actor's inputs.
• Proxy is off by default (useApifyProxy: false) and normally unnecessary — the backend is not behind Cloudflare/WAF.

Billing: pay-per-event — Listing $0.0009 per property row + Agency $0.0009 per agency row + Actor Start $0.00005 per run.

Legal & Compliance Disclaimer

This Actor is an independent project and is not affiliated with, endorsed by or sponsored by Zameen.com or its operators.

It accesses only publicly available, unauthenticated data that Zameen renders for anonymous visitors — plain-HTTP reads of public search results and property pages (window.state JSON and Elasticsearch _search responses). No browser, no login, no authentication or anti-bot bypass; the Actor does not touch /api/userEvents, /account, /nfpage/ or /captchaChallenge.

You are responsible for complying with Zameen.com's Terms of Service and applicable data-protection law — including Pakistan's PDPL and, where relevant, UAE PDPL — before commercial redistribution; respect robots.txt and rate-limit usage.

The output includes agent and agency phone numbers — business contact details published by Zameen for public viewing. They must not be used for unsolicited commercial outreach in violation of applicable law; this Actor targets B2B lead generation, not consumer cold-outreach.