Equipment Auction Comps — Heavy Machinery Price Monitor
Pricing
from $1.00 / 1,000 results
Equipment Auction Comps — Heavy Machinery Price Monitor
Track used construction and agricultural equipment listings across auctions and marketplaces in one schema, with first_seen so you can see what is newly listed and what has moved.
Pricing
from $1.00 / 1,000 results
Rating
0.0
(0)
Developer
alaudin burki
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
5 days ago
Last modified
Categories
Security
Limited permissionsShare
Equipment Auction & Used-Listing Comps — Heavy Machinery Price Monitor
Track used construction and agricultural equipment across marketplaces in one schema, and get the
comparison a dealer actually wants: for each make and model, how many are listed right now and at
what asking prices — plus firstSeenAt, so you can see what is newly listed and what has moved.
What you get
Rows — one per machine
| Field | Description |
|---|---|
id | Stable id, mascus:hi6l5rbp |
make | Manufacturer, resolved from a known-makes list — john deere, not john |
model | Model with the machine-type words stripped — 314e lcr |
title | Readable name, cat 314e lcr |
category | The category it was found under, e.g. crawler-excavators |
typeWords | The machine-type words found in the slug, e.g. tracked excavator |
price | Asking price if the index exposes one, else null — never guessed |
url | Direct link to the listing |
isNew | true when onlyNewSinceLastRun is on and this id was not seen before |
firstSeenAt | When this listing first appeared — carried across runs |
scrapedAt | This run's timestamp |
A second dataset, comps — one row per make + model
| Field | Description |
|---|---|
machine | cat 314e lcr |
listings | How many are listed right now |
priceLow / priceHigh | The asking-price range, when prices are exposed |
pricesSeen | How many listings actually carried a price |
categories, sources | Where they came from |
sampleUrl | One listing to look at |
Sample output
{"id": "mascus:hi6l5rbp","make": "cat","model": "314e lcr","title": "cat 314e lcr","category": "crawler-excavators","typeWords": "tracked excavator","price": null,"url": "https://www.mascus.com/construction/crawler-excavators/cat-314e-lcr-tracked-excavator/hi6l5rbp.html","isNew": null,"firstSeenAt": "2026-10-17T18:04:11.201Z","scrapedAt": "2026-10-17T18:04:11.201Z"}
{ "machine": "cat 930h", "make": "cat", "model": "930h", "listings": 3,"categories": "wheel-loaders", "sources": "mascus", "pricesSeen": 0 }
Default input
Runs with zero editing: three categories, Mascus, up to 1,000 listings, comps on. First run returns everything; turn on Only new since last run after that to get only what changed.
Limitations — stated plainly, because a surprise costs you a review
- One source today: Mascus. Its robots.txt is permissive and its listing URLs are a stable contract. It is the only source that answered.
- Ritchie Bros, MachineryTrader, AuctionTime and TractorHouse all return HTTP 403 to a plain request. They are not scraped here — faking them would give you silently empty rows. Covering them needs the platform's proxy or a browser session.
- Asking prices only, and only where the index exposes them. Mascus's category pages do not
publish a price, so
priceis usuallynull. Sold prices at auction live on individual lot pages and are not collected. Do not readpriceLowas a market value whenpricesSeenis 0. - Make and model are parsed from the URL slug against a known-makes list. An unfamiliar brand is split at the first word, so a new manufacturer can appear as a two-word make.
- This is not an appraisal tool. It reports what is listed and what is asked — never what a machine is worth.
Pricing
Pay per result — you are charged for the rows you receive, not for the run. Leave Max listings low
while you are testing.
Notes on how it behaves
- Every URL comes from the listing slug, never from generated CSS class names — Mascus's classes are hashed and change on deploy, and a parser built on them breaks silently.
- Zero listings across every category is treated as a failure, not an empty result, so a layout change surfaces as an error instead of a blank CSV.
compsis a separate named dataset, not extra rows in the main one.