Real Estate Listing Scraper & Market Stats

Scrape property listings from the public listing or search-result pages you supply — price, area, price per m², rooms, location, property type, agent and images — then get simple market statistics and a change report (new, changed and removed listings) against your previous run.

What does this real estate scraper do?

• Extracts one row per listing: title, price, currency, area (always m²; sq ft is converted), price per m², rooms, location, property type, agent, description, images and URL.
• Calculates market statistics per listing type and currency: median, minimum and maximum price and median price per m². Sale and rent, or USD and EUR, are never mixed. Outliers are excluded with a 1.5×IQR rule when there are at least 8 prices.
• Compares each run with your last run of the same URL and reports new, changed and removed listings — schedule it to monitor a neighbourhood or a portal search.
• Reads only the pages you give it. It does not crawl portals, log in or bypass bot protection.

Which sites work?

Listings are found with a generic HTML heuristic (cards that contain a price), not per-site parsers. It works on server-rendered pages such as onthemarket.com and rightmove.co.uk search results. Sites that show a bot check or render with JavaScript (for example realtor.com, hepsiemlak, idealista) are reported as errors in SUMMARY and add no dataset rows — and cost nothing.

Currency is read from the price text; $, EUR, GBP and TRY (TL) are recognised.

How to use it

1. Run a search on a supported portal and copy the results page URL.
2. Paste one or more URLs into Start URLs, set Max listings and optional Keywords (for example flat).
3. Click Start and export the dataset as JSON, CSV or Excel.
4. Schedule the same input to track new and removed listings over time.

Input example

{ "startUrls": [{ "url": "https://www.onthemarket.com/for-sale/property/london/" }], "maxListings": 30, "keywords": ["flat"] }

Output

• Dataset: only listing rows, billed per listing. Fields a page does not expose are null. Blocked, failed or empty sources add no rows.
• SUMMARY record (not billed): per source URL, new / changed / removed listings against your last run of that URL (the first run is baseline), or the error (invalid URL, private address, HTTP error, captcha or bot wall, JavaScript-only page).
• MARKET_SUMMARY record (not billed): median, min and max price and median price per m², one entry per listing type and currency.

Snapshots live in the named key-value store real-estate-snapshots.

How much does it cost?

You pay per listing delivered; the Pricing tab shows the current price. Blocked or empty pages cost nothing. Set a maximum total charge in the run options to cap spending.

FAQ

Can it scrape Zillow, Realtor.com or Idealista? Not reliably. Those sites render listings with JavaScript or block automated requests; the run reports them as errors instead of returning empty rows.

Is scraping real estate listings legal? It reads public pages only. Check each portal's terms of use, rate limits and the housing and privacy rules that apply to you before using the data.

Can I schedule it and use the API? Yes. Save the input as a task, schedule it, and connect results to Google Sheets, Slack, Make, Zapier or a webhook.