TractorHouse Scraper

Scrape TractorHouse.com — the leading online marketplace for tractors, farm equipment, construction machinery, trailers, and trucks. Search by keyword, category, manufacturer, price/year/hours range, US state or Canadian province, and condition — or fetch a full detail record (specs, seller contact, photos, video) for a specific listing ID or URL. No login, no cookies, no paid proxy.

What this actor does

• Two modes: search (keyword/filter browse) and byListingIds (full detail lookup)
• 51 equipment categories — tractors (by HP class), harvesters, tillage, hay & forage, dozers, excavators, skid steers, wheel loaders, telehandlers, trailers, trucks, and more
• 40 manufacturer filters — John Deere, Case IH, New Holland, Kubota, Massey Ferguson, Caterpillar, Bobcat, and more
• Location filter — every US state + Canadian province
• Price / model-year / operating-hours range filters
• Condition (New/Used) filter
• Full specs on detail lookups — every spec the seller listed (year, hours, drivetrain, tires, etc.), seller name/phone/location, all photos, and video if present
• Empty fields are omitted

Output per listing

Search mode

• listingId, title, category, manufacturer, model, description
• price, currency, condition, availability
• imageUrl, listingUrl
• locationState, locationPostalCode, locationCountry, locationStreetAddress (when public)
• recordType: "listing", scrapedAt

Detail lookup mode (byListingIds)

Everything above, plus:

• year, hours, listingType, priceDisplay, priceValidUntil
• specs — every spec/attribute the seller entered (varies by equipment type)
• sellerName, sellerPhone, sellerLocationCity, sellerLocationState, sellerLocationFull
• imageUrls[] (up to 20), videoUrl (when the listing has a video)
• recordType: "listingDetail", scrapedAt

Input

Field	Type	Default	Description
mode	select	search	search / byListingIds
keywords	string	–	Free-text search (mode=search). Leave blank to browse by category/manufacturer/state/etc. alone -- a non-blank value ANDs a keyword filter on top of any other filters set.
category	select	–	One of 51 equipment categories
manufacturer	select	–	One of 40 manufacturer brands
state	select	–	US state / Canadian province
condition	select	–	Used / New
priceMin / priceMax	int	–	Price range in USD
yearMin / yearMax	int	–	Model-year range
hoursMin / hoursMax	int	–	Operating-hours range
listingIds	array	–	Listing IDs or full listing URLs (mode=byListingIds)
maxItems	int	25	Hard cap (1–1000)

Example: John Deere tractors under $60,000 in Iowa

{
  "mode": "search",
  "category": "tractors",
  "manufacturer": "JOHN DEERE",
  "state": "IOWA",
  "priceMax": 60000,
  "maxItems": 50
}

Example: recent low-hour excavators

{
  "mode": "search",
  "category": "crawlerExcavators",
  "yearMin": 2020,
  "hoursMax": 1500,
  "condition": "Used",
  "maxItems": 50
}

Example: full detail lookup

{
  "mode": "byListingIds",
  "listingIds": [
    "257864499",
    "https://www.tractorhouse.com/listing/for-sale/241503499/some-listing-slug"
  ]
}

Use cases

• Equipment dealers — track competitor inventory and pricing by manufacturer/category/region
• Ag-tech & market intelligence — analyze used-equipment pricing trends by model year and hours
• Procurement — monitor new listings matching a specific make/model/price band
• Lead generation — pull seller contact details from detail lookups for outreach

FAQ

Do I need a TractorHouse account or API key? No. This actor scrapes publicly accessible listing pages — no login, cookies, or API key required.

Why are there two modes? search is fast and returns the browse-page fields (price, condition, location, thumbnail) for many listings at once. byListingIds fetches the full detail page per listing — richer (full specs, seller contact, photo gallery, video) but one HTTP request per listing, so use it for a shortlist rather than a full category sweep.

What does specs contain? TractorHouse lets sellers fill in category-specific attributes (e.g. tires vs. tracks for a skid steer, row spacing for a planter, drivetrain for a tractor). The actor passes through every spec key/value the listing has — the set of keys varies by equipment type.

Can I combine filters? Yes — category, manufacturer, state, condition, and all four range filters (price, year, hours) can be combined freely in search mode.

Why do search-mode records not include year/hours? TractorHouse's search-results cards don't expose those in their structured data — only the detail page does. Use yearMin/yearMax/hoursMin/hoursMax as server-side filters in search mode (they still narrow results correctly), or use byListingIds to get the explicit year/hours fields back per listing.

How fresh is the data? Listings are live at request time — TractorHouse is a real-time marketplace, not a static catalog.

Is this actor global? Coverage spans the US and Canada (all 50 states + 10 provinces are supported location filters); most listed equipment is located in North America.

Why might a very large maxItems return fewer records than requested? TractorHouse sits behind a Cloudflare managed challenge. The actor uses stealth-hardened Playwright with rotating proxy sessions and automatic retries per page, which reliably gets past the challenge for the first several search-result pages, but on a broad, unfiltered search Cloudflare can start rate-limiting further pagination mid-run. When that happens the actor fails soft — it keeps every listing already collected (no partial/corrupt records) and ends the run cleanly rather than erroring out. Narrowing the search (category/manufacturer/state) or re-running typically recovers full pagination.