Use case
Price monitoring from competitor and marketplace sites
Track the prices a store or marketplace shows for the products you care about, as records from an API instead of a scraper you maintain. Each call returns price, currency and availability for the products in scope.
The problem
Prices live on product pages, and the pages are built for shoppers, not for programs. A price tracker usually starts as a script that reads a few page elements. It then breaks when the store changes its layout, moves a price into a script tag or adds a sale badge next to the number.
Each competitor and marketplace needs its own version of that script, and each one needs a person to notice when it stops returning the right value. A wrong price that looks plausible is worse than an error, because nothing flags it.
How CleanedWeb solves it
A website becomes an API. You give CleanedWeb the product listing or category page and it does the work of turning it into records you can call.
-
Discover the fields
CleanedWeb reads the pages and proposes the fields it finds, such as title, price, currency and availability. You review them and keep the ones you need.
-
Build the API
It builds and validates an API for that site, with inputs such as a category or search term and a limit on how many records one call returns.
-
Get a versioned contract
The fields and their types are saved as a versioned schema. Your code can pin the version it was written against, so a changed contract shows up as a version conflict and not as a silently different price.
-
Run it on your schedule
Call the API from your own scheduler, hourly, daily or whatever your pricing decisions need, and store each response with its run details to compare prices over time.
-
Repairs when the site changes
When a site changes its layout, CleanedWeb rebuilds the API for it. A rebuild that would remove or retype a field is not swapped in silently. Your own APIs keep their contract until you review a change.
What a record looks like
An illustrative record from a fictional store. The fields of a real API come from the site you choose and are listed on its contract.
{
"sku": "LAMP-0420",
"title": "Desk lamp, matte black",
"price": { "amount": 49.0, "currency": "USD" },
"list_price": { "amount": 59.0, "currency": "USD" },
"availability": "in_stock",
"url": "https://shop.example.com/products/desk-lamp-matte-black",
"observed_at": "2026-01-15T09:00:00Z"
}
Store each response with the time you observed it. Comparing today's records with yesterday's gives you the price change, and the version of the contract tells you the fields still mean the same thing.
Docs for this use case
- Schedule API runs Run collection from your own scheduler with bounded, idempotent requests.
- Store and deduplicate Keep each response with the evidence needed to compare runs.
- Versions and changes Pin the contract your price comparison was written against.
- Source changes What happens when a store changes its pages.
Free tools that help
- Website data checker See whether a store already publishes JSON-LD, a feed or a sitemap before you build.
- Shopify store to CSV Take a one-off snapshot of a Shopify store's public prices and variants.
- JSON to CSV converter Turn a response into a spreadsheet to review prices by hand.
- JSON Schema generator Write the schema for the price records you store.
Questions
Which sites can I track?
Any public site whose product pages show the price without a login. Some sites block automated access or require sign-in; those cannot be turned into an API this way. Check a site with the website data checker, then start a build to see what CleanedWeb finds.
How often can I collect prices?
As often as you call the API. CleanedWeb does not run a scheduler for you; you call the API from your own cron job or workflow runner. Set the interval from how fresh the prices need to be and what each run may collect.
What happens when a store redesigns its pages?
A catalog API that keeps failing is rebuilt from the site as it is now, and the rebuild replaces the published version only after it passes checks. For an API you built, a change that would alter its contract waits for your review. See Source changes in the docs. A repair is not guaranteed, so check each run's summary.
Do I get historical prices?
No. Each call returns what the site shows now. To build a price history, store each response with its observation time.
Ready-made APIs in the catalog
These published APIs already return records for this kind of work. Open one to see its fields and a sample record.
-
Blocket Vehicle Listings API
Car listings from Blocket.se search results, with title, price, year, mileage, location, image and listing link.
-
Clas Ohlson Product Details API
One Clas Ohlson product page, read by its URL, with title, brand, SKU, price, currency, rating, review count, availability and link.
-
Craigslist Cars & Trucks (SF Bay Area) API
Car and truck listings from Craigslist SF Bay Area with title, price, location and listing link, searchable by keyword, make and model, body type, fuel…
-
DBA Marketplace Listings API
Second-hand listings from DBA.dk in Denmark, with title, description, date and category, plus price and brand from each ad's page.
-
ehorses Seller Details API
One seller profile on ehorses, read by its URL, with the seller's name and link, plus a phone number when the page shows one.
-
Kith Product Details API
One product page from kith.com, read by its URL, with title, price, currency, description and brand.
More use cases
-
Job data
Collect postings from job boards and career pages as records with title, employer, location and date.
-
AI data pipelines
Give RAG pipelines, agents and LLM apps records with a versioned schema instead of scraped page text.
-
Product catalog sync
Keep your catalog current with supplier and store sites using records you can diff and import.
Turn a site into an API
CleanedWeb turns a website into a maintained REST API that returns structured records, with a versioned schema and repairs when the site changes.