Yad2 Scraper

Yad2 Scraper: cars, property and second-hand goods from Israel's classifieds

Pick a section, add the filters you want or paste a Yad2 search URL, and you get one row per listing. Hebrew title, price in shekels, city and area, private seller or dealer, photos, the publish date and the link.

Yad2's vehicle and property feeds have no free-text search at all, so there is no keyword box for those. You filter by manufacturer, area, region, price and year instead. The second-hand marketplace is the opposite: it has no browse-everything route, so it needs a search term.

🔍 What Yad2 Scraper does

Three parts of Yad2 in one actor. Vehicles covers cars, motorcycles, scooters, trucks and watercraft. Real estate covers apartments and houses for sale, for rent, and commercial property. Second-hand goods is Yad2's marketplace.

Names work in Hebrew or English on both sides. טויוטה and Toyota resolve to the same manufacturer, and so do אזור חיפה והסביבה and Haifa area. 69 manufacturers and 43 areas are recognised, and an unknown name is reported rather than silently ignored.

Real estate is swept region by region, and each region is split into price bands so the crawl can reach past where Yad2 stops paginating one list.

📥 What you give it

{
  "section": "vehicles",
  "vehicleType": "cars",
  "manufacturers": ["Toyota"],
  "areas": ["Tel Aviv-Yafo area"],
  "minYear": 2018,
  "maxPrice": 120000,
  "maxItems": 500
}

sellerType is applied after the results come back, because Yad2 has no seller filter of its own. Listings that do not match are dropped and never charged. On vehicles the crawl also stops once a full round of results contains none of your seller type, so a narrow filter can return fewer than maxItems while more matches exist further down.

Rows are written at the end of the run, after the detail pass. A very large maxItems with details on can hit the run timeout before anything is delivered. Split big pulls across runs, or turn includeDetails off.

A cityId with no regions sweeps all eight to find the right one. You will see a handful of NO_RESULTS rows next to your real results. They are not charged.

📤 What you get back

A real car listing from a recent run, with most of the photo URLs cut:

{
  "listingId": "bdevnzns",
  "section": "vehicles",
  "subSection": "cars",
  "title": "טויוטה RAV4 Limited היברידי אוט׳ 2.5 (155 כ״ס) 2017",
  "yad2Title": "טויוטה RAV4 שנת 2017 יד שלישית 204,000 ק\"מ",
  "contactName": "גל",
  "description": "טויוטה ראב4 ההיברידית, רכב אמין במיוחד עם חיסכון בדלק...",
  "price": 82900,
  "currency": "ILS",
  "previousPrice": null,
  "sellerType": "business",
  "sellerName": "אוטו אדיר",
  "sellerId": "5263113",
  "categoryId": 1,
  "subCategoryId": 21,
  "city": "פתח תקווה",
  "area": "פתח תקווה והסביבה",
  "areaId": 4,
  "region": null,
  "neighborhood": null,
  "street": null,
  "latitude": null,
  "longitude": null,
  "publishedAt": "2026-07-02T12:48:55+03:00",
  "updatedAt": "2026-09-14T08:45:58+03:00",
  "manufacturer": "טויוטה",
  "model": "RAV4",
  "subModel": "Limited היברידי אוט׳ 2.5 (155 כ״ס)",
  "year": 2017,
  "hand": "יד שלישית",
  "engineType": "היברידי בנזין",
  "engineVolume": 2494,
  "kilometers": 204000,
  "rooms": null,
  "squareMeters": null,
  "floor": null,
  "propertyType": null,
  "condition": null,
  "tags": ["גלגלי מגנזיום", "בקרת שמירת מרחק", "בקרת שיוט אדפטיבית"],
  "images": ["https://images.yad2.co.il/Pic/202607/02/1_3/o/y2_1pa_010458_...", "..."],
  "imageCount": 8,
  "adNumber": "57171185",
  "url": "https://www.yad2.co.il/vehicles/item/bdevnzns",
  "searchUrl": null,
  "scrapedAt": "2026-09-14T05:48:04.651Z"
}

Every section returns the same columns, and the ones that do not apply are null. A flat has no odometer reading and a car has no room count, and both fit one table.

🧾 Reading the output

Three kinds of row can land in your dataset.

A real listing row carries no charged field, so filter on _sample and _diagnostic being absent. Run it with no section and no URL and you get the sample row on its own.

If a detail request fails for one listing, the row is still returned and still charged, with publishedAt, description and yad2Title left null and nothing on the row to mark it. Check for a null publish date on vehicle and property rows if that field matters to you.

▶️ How to run it

1. Open Yad2 Scraper and click Try for free. 2. Pick a Section. For second-hand goods, type a Hebrew Search term. 3. Add filters: manufacturer and area for vehicles, region or city id for property, price for both. 4. Set Maximum listings. Start around 100 to see the output shape, then click Start. 5. Download the dataset as JSON, CSV or Excel, or read it from the Apify API.

💰 How much does it cost?

$1.20 per 1,000 listings. Flat on every Apify plan, no volume tiers.

You pay per listing row delivered. The sample row, diagnostic rows, duplicates and listings dropped by the seller-type filter carry no listing charge, and a search that matches nothing produces none either. Turning on the detail pass costs nothing extra per listing.

💡 What people use it for

• Watching one manufacturer and area daily and picking up what was listed since yesterday, deduping

on listingId.

• Building a price curve for a model by year and mileage across a few thousand rows.
• Pulling a rental region with sellerType on private to skip the agencies.
• Tracking how long listings stay up, by comparing publishedAt against when they stop appearing.

🚧 What it does not do

• No contact details. Yad2 does not publish phone numbers or emails, so neither does this.

contactName and sellerName are as far as it goes.

• No keyword search on vehicles or property. Yad2's own feeds have none. Use the filters.
• Second-hand goods need a search term and come back without a publish date, because that feed

does not carry one.

• Real estate is deep, not complete. A region reaches a few thousand listings through the price

band splitting, not everything Yad2 holds.

• Roommates and agricultural machinery are not covered.
• Second-hand results are not repeatable. Yad2 orders that marketplace differently on each

request, so two identical runs return overlapping rather than identical sets.

• A pasted URL wins. The form's year, kilometre, manufacturer and area fields are ignored for

that search.

• Rows are a snapshot. A listing can be edited, sold or removed afterwards, and scrapedAt

records when it was read.

🧭 Which classifieds scraper do you need?

❓ Questions people ask

Do I need Hebrew? Not for the filters. Manufacturer and area names work in English. The second-hand search term does better in Hebrew, because that is what sellers write.

Where do I find a city id? Open the city on Yad2 and read the city= value out of the URL. 5000 is Tel Aviv-Yafo, 3000 Jerusalem, 4000 Haifa.

Why is publishedAt empty? Either the listing is second-hand goods, which carry no date, or you turned the detail pass off, or that one detail request did not come back.

Can I filter to private sellers? Yes, set Seller type. It is applied after the results arrive, and the dropped rows are not charged.

How many listings can I get? Up to 50,000 in one run. A single vehicle search reaches about 59,000, so your cap is usually the limit rather than Yad2.

Is scraping classifieds legal? These are public listings on public pages. They carry personal data all the same, which GDPR and similar laws cover, so have a reason for collecting it. Apify's write-up on scraping and the law is a good starting point, and we are not lawyers.

🆘 If something breaks

Open the Issues tab on the actor page. Send the section and filters you used, or the URL, and the run ID. The errorCode on the diagnostic row usually names the problem on its own.