Agriaffaires: Used Farm Machinery Classifieds Data

One row per machine advertised for sale across the largest farm-machinery classifieds network in Europe. Agriaffaires sits inside the Groupe BV portfolio of vertical classifieds brands - the same stable as MachineryZone for construction-adjacent iron and Trucks Corner for commercial vehicles - and runs localized editions in 18+ countries: France, Italy, Spain, Germany, the Netherlands, Poland, Portugal, Romania, the Czech Republic, Bulgaria, the United Kingdom, Canada, the USA, Brazil, China and Ukraine among them.

At research time the UK edition advertised 29,167 used-tractor ads, paginated at 50 per page across roughly 200 result pages - and tractors are only one aisle. The catalogue spans harvesting equipment, seed drills, sprayers, tillage and power harrows, spreaders, hay and forage kit, farm trailers, irrigation, livestock equipment, spare parts and even damaged or burned machines sold for salvage.

What turns the marketplace into a dataset is the card schema underneath the browsing experience. Every ad exposes make and model, a numeric ad ID, year of manufacture, engine hours, horsepower, an asking price anchored to a canonical euro reference value, the equipment category, a photo count and the seller's trading name with country and region. Ten typed fields, consistent across editions, which is exactly what makes cross-border price comparison computable rather than anecdotal. Get a sample of this dataset and the rows arrive cut to whichever makes, categories and regions you sell into.

Flat, tidy, one row per advertised machine - shaped to load straight into a warehouse table. Five real rows in the delivered column order:

title: John Deere 8RX410    year: 2020  hours: 6570   hp: 410  price_eur: 154061   seller: Ben Burgess & Co (United Kingdom, East Midlands)
----------------------------------------------------------------------------------------------------------------------
title: John Deere 6R185     year: 2025  hours: 991    hp: 185  price_eur: 136554   seller: Ben Burgess & Co (United Kingdom, East Midlands)
----------------------------------------------------------------------------------------------------------------------
title: John Deere 2750      year: 2019  hours: 1970   hp: -    price_eur: 20950    seller: Hunt Forest Group (United Kingdom, South East England)
----------------------------------------------------------------------------------------------------------------------
title: John Deere 6155M     year: 2023  hours: 1875   hp: 155  price_eur: 91911    seller: Farol Ltd (United Kingdom, South East England)
----------------------------------------------------------------------------------------------------------------------
title: John Deere 8370R     year: 2019  hours: 4777   hp: 370  price_eur: 145891   seller: Hunt Forest Group (United Kingdom, South East England)

Read what the block already says. The 8RX410 and the 6R185 bracket the late-model large-frame market from the same Midlands dealership at 410 hp and 185 hp respectively; the 2750 is a 2019 workhorse at barely 2,000 hours and an eighth of the flagship's price, with horsepower unstated on the card exactly as the marketplace leaves it; the pair of Hunt Forest Group listings shows how one dealer group populates multiple price-and-hour cohorts at once. Every figure arrives typed, not parsed out of prose. Request a sample and you get rows in your own makes and regions, populated exactly as shown.

Ten verified fields define the ad record: four describe the machine, three describe the sale and three anchor it to the seller. Definitions were assembled from the attributes every listing card carries, so the dictionary below is the confirmed core - anything beyond it sits behind the request note rather than being padded with invented columns.

Geography - Europe first and foremost, through localized editions in France, Italy, Spain, Germany, the Netherlands, Poland, Portugal, Romania, the Czech Republic, Bulgaria and the United Kingdom, extended to Canada, the USA, Brazil, China and Ukraine - 18+ editions carrying the same card schema. Because every price anchors to a canonical euro value, a same-model comparison between a Midlands dealer and a Lombardy one needs no currency reconciliation before it starts.

Temporal - live classifieds. Ads appear when a machine hits the yard and expire when it sells, so a snapshot reads as genuine current supply rather than a stale catalog. There is no published historical archive: trend work accrues forward from your first pull, and repeated snapshots trace turnover - which machines moved, how fast, and at what asking price.

Granularity - one record per machine ad, not per dealer rollup or monthly average. The UK used-tractor pool alone ran to roughly 200 result pages of 50 ads at research time. Ad-level grain is where price discovery happens: individual machines with individual hours, individually priced, individually negotiable.

API, files, or your warehouse. Daily, weekly, or hourly.

Your cadence is decoupled from how fast the classifieds turn over. Take a full snapshot once to baseline your categories, or keep a warehouse table current so newly posted ads and sold-listing removals diff cleanly into yesterday's rows. Deliveries arrive normalised to the field dictionary above, with the euro-canonical price preserved as its own column so multi-edition comparisons stay exact.

• Used-equipment valuation and price discovery - fit asking-price curves by make, model, year and hour band across a whole continent, then benchmark any single machine against its cohort before a trade-in, an appraisal or a fleet disposal.

• Depreciation and residual modeling - hours against price across thousands of same-model tractor ads gives lenders, lessors and OEM finance arms an empirical depreciation surface instead of a rule-of-thumb percentage.

• Dealer competitive intelligence - count live inventory by dealership, region and category to see who is stocking what, where used supply is thinning, and how quickly rival yards turn their iron.

• Cross-border sourcing and arbitrage - the euro-canonical price column makes same-model gaps between country editions visible without an FX layer, which is the entire basis of cross-border machine sourcing.

• Aftermarket sizing - the age-band distribution of the live fleet by horsepower class is the denominator parts and service planners usually have to guess at.

• Machine-learning training sets - a tidy ten-field table with typed numerics trains price-prediction and residual models without image or text augmentation.

Market researchers and consultants get ad-level supply and pricing for the European used-machinery market instead of trade-press anecdotes; see market researchers use cases. Sales and growth teams at dealerships and OEM branches get competitor inventory depth by yard, region and category; see sales growth teams use cases. Investors and quants get a live pricing panel for equipment-heavy ag businesses and a depreciation input most residual tables lack; see investors quants use cases. Data scientists and ML engineers get a typed, join-ready ten-field table that trains valuation models out of the box; see data scientists use cases.

Within agricultural and farm machinery data, this record owns the live offer side of the European used market. Its closest neighbour, MachineFinder, is John Deere's own used-equipment marketplace - deep on one manufacturer at roughly 77,500 machines, but silent on New Holland, Fendt, Case IH and every other marquee in the yard. Pick MachineFinder for single-brand trade-in benchmarking; pick Agriaffaires for the whole multi-marquee European market. At the other pole, FAOSTAT's agricultural machinery series counts tractors and harvesters in use across roughly 250 countries back to 1961 - a macro census of the installed base, not a price discovery panel. If your question is "what is this machine worth, today, in this market", this is the record; if it is "how large is the world's tractor fleet", take FAOSTAT.

Three things worth knowing upfront. First, this is the ask side: prices are what dealers and private sellers advertise, not hammer prices. They are the right input for supply, valuation and residual work, and pair naturally with auction-results data when a transaction-grade number matters.

Second, the corpus is current-state: ads expire when machines sell and there is no published archive, so longitudinal studies start accruing from your first delivery. Repeated snapshots are how turnover gets measured.

Third, the ten fields above are the confirmed core; sub-type classifications, damaged-machine and spare-parts designations and currency-display variants confirm with your sample rather than appearing here speculatively. Name your makes, categories and regions and the sample returns in exactly the schema shown above.