Tractor Junction Scraper: Used and New Tractors, Harvesters and Dealers in India

Tractor Junction Scraper is an Apify Actor that extracts tractor and farm machinery data from tractorjunction.com, the Indian marketplace for tractors, harvesters and implements. It returns used ads (13,183 used tractors, 959 used harvesters and 3,238 used implements on 2026-10-23), 876 new tractor models with specifications and prices, new harvester and implement models, and 1,418 certified dealers. Prices are in Indian rupees. No login, proxy or browser is needed.

Export the result as JSON, CSV, Excel, XML or HTML, or read it through the Apify API.

What can you scrape from Tractor Junction?

• Used tractors: the national list or any mix of state, brand, price band, model year and horsepower band. Each row has price in INR, monthly EMI, make, model, year, city, district, state, photos and the Certified badge.
• Used harvesters and used implements: the same list rows with price, year, state, city and photos, plus the working-hours band for harvesters. Filter by state, brand or implement category.
• Ad details (optional): horsepower, tyre and engine condition (percent bands), RC availability, financier NOC, the state and office code of the registration, the site's reference price, the full description and the inspection checklist of certified tractors. Harvester and implement details add power source, cutter bar width, crop, implement type, the listing date and the Sold flag.
• New tractors, harvesters and implements: one row per model with minimum and maximum price, horsepower, cylinders, engine capacity, torque, drive, gearbox, weight, lift capacity, warranty, rating, review count and the complete specification tables grouped as the site shows them.
• Certified dealers: name, brand, address, city, state, pin code, opening hours and days.
• Monitoring: a scheduled run returns only rows you have not collected before, plus a price-change row when a known ad or model changes its price.

Why use this actor?

• Filters the site applies for you: used tractors take brand, state, price, year, horsepower and certified filters in one request, so only matching ads are read.
• Every record type in one place: ads, new models and dealers share one schema with stable column order.
• Clean, typed output: numbers are numbers, text is UTF-8 (Hindi seller notes stay intact) and fields that do not apply to a row are left out, never filled with null.
• Honest errors: an unknown brand, state or category ends with one free error row that lists the valid names, never another place.
• Privacy built in: seller names, phone numbers, e-mails and full registration plates are never collected.

How to scrape Tractor Junction in four steps

1. Create a free account at console.apify.com and open the actor.
2. Choose What to collect: used tractors, used harvesters, used implements, new tractors, new harvesters, new implements, dealers or addresses.
3. Add states, brands or categories and any price, year or horsepower filter, then set Maximum items.
4. Click Start, then download the dataset from the Storage tab.

A minimal input that returns certified used Mahindra tractors in Rajasthan, priced from 3 to 6 lakh rupees:

{
  "mode": "used-tractors",
  "states": ["Rajasthan"],
  "brands": ["Mahindra"],
  "certifiedOnly": true,
  "priceFrom": 300000,
  "priceTo": 600000,
  "maxItems": 50
}

Brand and state names are matched as the site writes them and case does not matter (Odisha and Orissa both work). Used tractor filters are combined by the site; used harvester and implement lists filter by one address at a time, so the actor applies the other filters to the cards it reads. New-model filters on horsepower and price read the model pages first, so a narrow range reads more pages than it returns.

What data do you get?

Every row has a listingType (used, new or dealer). The dataset schema has 70 fields.

Field	Description
listingId, referenceCode, url, title	Ad number or model path, the site's reference (TJN...), address and title
sourceSite, country, listingType, vehicleType	tractorjunction.com, IN, used, new or dealer, and tractor, harvester or implement
make, model, year, condition	Brand, model, year of purchase or model, Used or New
isCertified, isSold	Certified badge (tractors) and Sold flag (details of harvesters and implements)
price, priceMax, currency, monthlyPayment, comparisonPrice	Asking price in INR, top of the price range (new models), INR, EMI per month, the site's reference price
previousPrice, priceChange, priceChangePct, changeType	Price-change rows: old price, difference, percent, price-drop or price-increase
powerHp, ptoHp, engineSize, cylinders, torqueNm	Horsepower, PTO horsepower, engine capacity in cc, cylinders, torque
driveType, transmission, gears, topSpeedKmph, weightKg, fuelTankLitres, liftCapacityKg	Drivetrain and mechanical data of new models
powerSource, cutterBarWidth, crop, crops, hours	Harvesters: power source, cutter bar, crops, working-hours band
implementType, category, powerRequirement	Implements: kind, group and tractor power needed
tyreCondition, engineCondition, financierNoc, rcAvailable, rtoState, rtoOffice	Used tractor details: condition bands, financier NOC, RC, registration state and office code
warranty, modelStatus, rating, reviewCount	New models: warranty, Launched or upcoming, rating and number of reviews
city, district, region, postalCode, address	Location of the ad or dealer (region is the state)
openingHours, opensAt, closesAt, openingDays	Dealer hours
description, sellerNotes, inspectionChecklist, specifications	Site description, the seller's short note, certified inspection list, grouped specification tables
images, publishedAt, scrapedAt, error	Photo addresses, listing date (harvester and implement details), time of the read, error text on a free error row

Sample output

A real used-tractor row from a list read on 2026-10-23:

{
  "title": "Mahindra 585 DI XP Plus",
  "listingId": "261501",
  "url": "https://www.tractorjunction.com/used-tractor/mahindra/585-di-xp-plus-245428/261501/",
  "sourceSite": "tractorjunction.com",
  "country": "IN",
  "listingType": "used",
  "vehicleType": "tractor",
  "isCertified": true,
  "make": "Mahindra",
  "model": "585 DI XP Plus",
  "year": 2021,
  "condition": "Used",
  "price": 460000,
  "currency": "INR",
  "monthlyPayment": 9849,
  "city": "Bhadra",
  "district": "Hanumangarh",
  "region": "Rajasthan",
  "images": [
    "https://assets.tractorjunction.com/tractor-junction/assets/images/images/oldTractor/585-di-xp-plus-245428-1792561969-0.webp"
  ],
  "scrapedAt": "2026-10-23T14:39:07.601Z"
}

A new-model row for the Mahindra 475 DI shows a price of 649,000 to 679,000 INR, 42 HP, a 4-cylinder 2,731 cc engine, 194 Nm of torque, 8 forward and 2 reverse gears, a lift capacity of 1,700 kg and a rating of 4.9 from 94 reviews, followed by the full specification tables.

Monitor new ads and repricing

Switch on Only new rows (monitor) and the actor remembers every row it delivered, in a named store of your own account. Scheduled runs with the same Monitor name return only rows it has not delivered before. Add Report price changes and an ad or model whose price moved comes back as a row with the previous price, the difference, the percent and the change type. The site publishes no price history, so this is the way to build one. Skipped rows are never charged.

How does it compare with other tractor marketplace scrapers?

No other Apify actor targets tractorjunction.com or an Indian tractor marketplace (Store search on 2026-10-23). The tractor and farm machinery actors that exist cover US, European and Latin American sites, so this actor adds India: used tractor ads priced in rupees with state-level filters, certified tractors, and new-model specifications for the Indian market (Mahindra, Swaraj, Sonalika, Eicher, Massey Ferguson and others).

Is it legal to scrape Tractor Junction? Limits and responsible use

The actor reads public pages only. It does not log in, solve captchas or bypass protection, and it skips every path that the robots.txt of the site disallows (/ajax/, /api/, /admin/ and others). It paces requests at about one every 1.3 seconds, because the site's Cloudflare layer answers HTTP 429 above roughly one request per second, and it backs off when asked. Seller names, phone numbers, e-mails, reviewer names and full registration plates are never collected. The site's terms of use list scraping or harvesting data without authorisation as prohibited conduct, so check the terms and Indian data protection law before commercial reuse; you are responsible for how you use the data.

Limits:

• Used lists carry a shorter record than the ad page. Switch on Read each ad page (details) for horsepower, condition bands, reference number, description and the inspection checklist.
• Many used ads are old or already sold; the Sold flag is only visible on harvester and implement details.
• Dealer phone numbers and e-mails are masked by the site, so dealer rows carry business address and hours only.
• Prices and descriptions are what sellers typed.
• Free Apify plans are limited to 10 results per run.

Use with AI agents (MCP)

Connect the Apify MCP server to Claude, Cursor or another MCP client and this actor appears as a tool: https://mcp.apify.com?tools=RecordsData/tractorjunction-scraper.

Example prompt: "List 20 certified used Mahindra tractors in Rajasthan between 3 and 6 lakh rupees, with model year and EMI." The agent sends { "mode": "used-tractors", "states": ["Rajasthan"], "brands": ["Mahindra"], "certifiedOnly": true, "priceFrom": 300000, "priceTo": 600000, "maxItems": 20 } and analyzes the rows.

Run it from code

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_API_TOKEN' });
const run = await client.actor('RecordsData/tractorjunction-scraper').call({
    mode: 'new-tractors',
    brands: ['Mahindra'],
    powerFrom: 40,
    powerTo: 50,
    maxItems: 20,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items.length, items[0].title, items[0].price);

You can also call the actor from the Python client, Make, Zapier or n8n, and schedule it from the Apify Console.

Frequently Asked Questions

What does the Tractor Junction Scraper cover?

It covers the public data of tractorjunction.com: used tractor, harvester and implement ads from every Indian state, new tractor, harvester and implement models with specifications and prices, and certified dealers. It also has a monitor mode that returns only new rows and price changes on scheduled runs.

Do I need a proxy, a login or an API key?

No. The site answers plain HTTP requests from the Apify cloud, so the actor uses no proxy, browser or Tractor Junction account. You only need an Apify account to start runs, and the free plan returns up to 10 results per run. A proxy field exists only for networks that require one.

Is it legal to scrape Tractor Junction?

The actor reads publicly visible pages, does not log in and skips the paths that robots.txt disallows. It never collects seller names, phone numbers, e-mails or full plates. The site's terms of use list scraping without authorisation as prohibited conduct, so you remain responsible for your use under those terms and Indian data protection law.

How many used tractors can I get?

On 2026-10-23 the national used-tractor list had 13,183 ads on 1,648 pages of 8, with no depth cap. Used harvesters numbered 959 and used implements 3,238. Leave states and brands empty for all of them and set Maximum items to the number you need.

How much does a run cost?

The actor is billed per result, with one rate for each record type: list ad, ad with details, new model, dealer and price change. The Pricing tab of the actor page shows the live rate. Error rows, empty searches and rows skipped by monitoring are never charged.

How long does a run take?

The actor reads one page about every 1.3 seconds, because the site's Cloudflare layer blocks faster clients. In the cloud at 256 MB, runs of 3 used ads finished in 2 to 4 seconds, and 3 ads with details in 5 to 10 seconds. Searches that must open model pages to apply a filter take longer: 3 new tractors under 600,000 rupees took 38 seconds and 3 Fieldking rotavators about 33 seconds, because the actor read the model pages one at a time until it found three. A filter that matches nothing reads the whole list (876 tractor models, 119 harvester models, about 1,125 implement models), so keep Maximum items and the filters realistic.

Why do some used rows have no horsepower or reference number?

The list carries a shorter record than the ad page. Horsepower, tyre and engine condition, the reference number, the description and the inspection checklist live on the ad page. Switch on Read each ad page (details) to get them; that costs one more page read per ad and is billed as the detail record.

Can I track price changes?

Yes. Turn on Only new rows and Also report price changes, give the search a Monitor name and schedule the actor. A known ad or model whose price changed comes back with the previous price, the difference and the percent. The site publishes no price history, so the monitor builds one from your own runs.

What happens when the search finds nothing or the input is wrong?

A search with no hits ends as succeeded with an empty dataset and charges nothing. A wrong value, such as an unknown brand or state, an unknown implement category or an inverted price range, ends as succeeded with one free error row that explains the problem and lists the valid names, so a typo never costs you money.

Support

Questions, a missing field or a feature request: contact.punkrecordsdata@gmail.com.