By the end of this tutorial you will have a Python script that pulls Belgian property listings from immoweb.be into structured rows: asking price, address, living area, land area, bedrooms, EPC label, agency and photo links.
Immoweb is the Belgian real-estate portal. Immoweb Scraper (reventadata/immoweb-scraper) is the Apify Actor that reads its public listings. You describe a search with a few inputs, and it returns one row per listing. It is not affiliated with Immoweb.
This guide runs the Actor from Python with the Apify API client. The sample rows below come from a real run of a postal code 9000 search (Gent), copied from its dataset, not from an invented example.
Prerequisites
You need three things:
- Python 3.11 or newer. The Apify client for Python 3.x requires it.
- A free Apify account, created with an email address or a GitHub login.
- Your API token, from Settings > API & Integrations in Apify Console.
Install the client and export the token:
pip install -U "apify-client~=3.0"
export APIFY_API_TOKEN="your_apify_token_here"
No scraping code is needed on your side. The Actor reads public pages, with no login, and does the fetching.
What you'll achieve
You will search Gent (postal code 9000) for houses for sale, read the rows in Python and print a summary line per listing. The summary includes a price per square metre that you work out yourself from two fields. The example run finished in 4.2 seconds.
The same script covers a rental search, another postal code or a price ceiling. Only the input changes. Each run takes one propertyType and one transaction, so a mixed search of houses and apartments needs one run per type.
Step 1: Open Immoweb Scraper
Open the Immoweb Scraper page on Apify Store. The Try for free button there opens the Actor in Apify Console. This tutorial calls the Actor from the API, so you only need its name, reventadata/immoweb-scraper.
Step 2: Choose the search
The search is described by these input fields:
-
propertyType:houseorapartment. Any other value is refused before the run starts. -
transaction:buyorrent. On a rent search,priceis the monthly rent. -
postalCodes: a list of four-digit Belgian postal codes such as9000(Gent) or1000(Brussels). Each code is one search, run in the order given. An empty list searches all of Belgium in one search. -
minPriceandmaxPrice: optional limits in EUR. An empty field means no limit. On a rent search they limit the monthly rent. -
maxItems: a cap on rows across all searches. The default is 1000. -
maxPages: an optional cap on result pages per search, with 60 listings on each page. -
fetchDetails: off by default. It adds the agency's public contact fields (see Step 5).
Immoweb pages any one search to at most 9,960 listings (166 pages of 60). A larger search is reported in the run log, and splitting it by postal code or by a price band reaches the rest.
To search two postal codes for apartments to rent, the input would look like this:
{
"propertyType": "apartment",
"transaction": "rent",
"postalCodes": ["9000", "1000"],
"maxPrice": 1200,
"maxItems": 50,
"maxPages": 2
}
Step 3: Run it from Python
This script starts the Actor with a search for houses for sale in postal code 9000, waits for it to finish, reads the dataset and prints two lines per listing. The first two input values are the Actor's defaults, so they could be left out.
import os
from apify_client import ApifyClient
# 1. Initialise the client with your token from the environment
client = ApifyClient(os.environ["APIFY_API_TOKEN"])
# 2. Describe the search: houses for sale in postal code 9000, ten rows
run_input = {
"propertyType": "house",
"transaction": "buy",
"postalCodes": ["9000"],
"maxItems": 10,
"maxPages": 1,
}
# 3. Start the Actor and wait for it to finish
run = client.actor("reventadata/immoweb-scraper").call(run_input=run_input)
if run is None or run.status != "SUCCEEDED":
status = run.status if run else "unknown"
raise SystemExit(f"Run did not succeed (status: {status}).")
# 4. Read the rows from the run's default dataset
items = client.dataset(run.default_dataset_id).list_items().items
print(f"Retrieved {len(items)} listings.\n")
# 5. Print two lines per listing, with a price per square metre worked out here
for item in items:
street = " ".join(part for part in (item.get("street"), item.get("houseNumber")) if part)
price, area = item.get("price"), item.get("livingAreaM2")
per_m2 = f"{price / area:,.0f} EUR/m2" if price and area else "no living area"
print(f"{street}, {item.get('locality')} | {price} EUR | {area} m2 | {per_m2}")
print(f" {item.get('propertySubtype')}, bedrooms: {item.get('bedrooms')}, EPC: {item.get('epcLabel')}")
The run reads its results page by page, so keep maxPages small while you test.
Step 4: Read the output
The loop prints one block per listing. Here are the first three listings of the example run's ten rows:
Sint-Denijslaan 359, Gent | 949000 EUR | 188 m2 | 5,048 EUR/m2
VILLA, bedrooms: 3, EPC: C
Wijmenstraat 16, Gent | 248000 EUR | 162 m2 | 1,531 EUR/m2
HOUSE, bedrooms: 3, EPC: F
Koningin Elisabethlaan 76, Gent | 1900000 EUR | 558 m2 | 3,405 EUR/m2
HOUSE, bedrooms: 3, EPC: B
Each row carries the full set of fields. One complete row from the same run, the second listing, follows. imageUrls is cut to its first entry, and agencyLogoUrl and searchUrl are left out.
{
"id": "21993768",
"url": "https://www.immoweb.be/en/classified/house/for-sale/gent/9000/21993768",
"title": "Home with an open-plan feel and a large garden",
"propertyType": "HOUSE",
"propertySubtype": "HOUSE",
"transaction": "FOR_SALE",
"price": 248000,
"currency": "EUR",
"priceLabel": "€248,000",
"priceKind": "residential_sale",
"priceMin": null,
"priceMax": null,
"previousPrice": null,
"monthlyCharges": null,
"bedrooms": 3,
"rooms": null,
"livingAreaM2": 162,
"landAreaM2": 875,
"epcLabel": "F",
"street": "Wijmenstraat",
"houseNumber": "16",
"box": null,
"postalCode": "9000",
"locality": "Gent",
"district": "Gent",
"province": "East Flanders",
"region": "Flanders",
"latitude": 51.0866374,
"longitude": 3.6650255,
"agencyName": "Cornelis Goeman Merelbeke",
"isProject": false,
"projectName": null,
"listingFlag": "new",
"hasVideo": false,
"hasVirtualTour": false,
"lastModifiedAt": "2026-11-16T16:00:20.044Z",
"imageUrls": [
"https://media-resize.immowebstatic.be/classifieds/6ec69a41-f934-4ba7-8e15-cb6597c88d5b/736x736/357beab8464da3e96487a2f79c340ae5.jpg"
],
"agencyId": null,
"agencyType": null,
"agencyPhone": null,
"agencyMobile": null,
"agencyEmail": null,
"agencyWebsite": null,
"contactDetailsFetched": false
}
Step 5: Add agency contact (optional)
Set fetchDetails to true and the Actor also opens each kept listing's own page. It adds the publishing agency's public phone, mobile, email and website to the row, in agencyPhone, agencyMobile, agencyEmail and agencyWebsite, with agencyId and agencyType. contactDetailsFetched shows whether the page was read.
The run is much slower with it on, because every listing's page is read as well. A listing whose page cannot be read still ships, with its contact fields empty and contactDetailsFetched set to false.
Understanding your results
Every row carries the fields below. Money is in EUR.
-
priceandpriceLabel: the asking price as a number and as shown on the listing.priceMinandpriceMaxcarry the range the listing shows, when it shows one.isProjectsays whether the listing belongs to a project. -
propertyTypeandpropertySubtype: Immoweb's category, such asHOUSEwith subtypeVILLA.transactionreadsFOR_SALEorFOR_RENT. -
street,houseNumber,box,postalCode,locality,district,provinceandregion: the address, split up. -
livingAreaM2,landAreaM2,bedroomsandrooms: sizes and counts, as published. -
epcLabel: the energy performance label, as published. -
agencyNameand the contact fields from Step 5. -
imageUrls,lastModifiedAtandlistingFlag: photos, the date of the listing's last change, and the flag Immoweb sets on it.
The ten rows of the example run show these things:
- Every row is a
HOUSEtype listing for sale in postal code 9000, with locality Gent. - The subtype is
HOUSEon 7 rows,VILLAon 1,MANSIONon 1 andAPARTMENT_BLOCKon 1. - EPC labels are C on 4 rows, B on 2, D on 2, A on 1 and F on 1.
- Bedrooms are 3 on 7 rows, and 2, 4 and 6 on one row each.
- Asking prices run from 248,000 to 1,900,000 EUR.
- Carlo Eggermont is the agency on 4 rows and Immo Da Vinci on 2. Vastgoedkantoor IMMO-HOME, Cornelis Goeman Merelbeke, Wolff Real Estate and LE CLAIR are on one row each.
- One row has an empty
title. Four titles end in "...". Titles are in Dutch and in English. -
box,monthlyCharges,priceMin,priceMax,previousPrice,projectName,agencyId,agencyType,agencyPhone,agencyMobile,agencyEmailandagencyWebsiteare empty on every row.contactDetailsFetchedis false on every row. -
listingFlagisnewon every row.isProject,hasVideoandhasVirtualTourare false on every row. - The
lastModifiedAtdates fall between 15 and 17 November 2026.
Pro tips
-
Start small. The example run used
maxItems10 andmaxPages1. Raise them once the rows look right. -
Split large searches. A search with more matches than Immoweb pages is reported in the log. Split it with
postalCodes, or with aminPriceandmaxPriceband, to reach the rest. -
Compute your own ratios. There is no price-per-square-metre field. Divide
pricebylivingAreaM2yourself, as the script does, and skip rows where either isnull. -
Keep rent and charges apart. On a rent search,
priceis the monthly rent andmonthlyChargesis the charges shown beside it. Add them only when you want the total. -
Use
fetchDetailsonly when you need agency contacts. It makes the run much slower.
Troubleshooting
- Zero rows, status SUCCEEDED. The site answered and no listing matched. The run's status message says so.
- Zero rows, status FAILED. A postal code that Immoweb does not know is skipped with an error in the log. When no row was saved, the run ends as FAILED with a message to fix the input. Pages that could not be read also end a run that saved no rows as FAILED, and the message gives the reason.
- Fewer rows than you expected. If the run stopped on its run-time limit, the rows already saved are kept and the log says so.
Next steps
For current pricing and details, see the Store page.
To compare another area, change postalCodes and run the script again, then sort the two sets of rows by price.
What postal code would you search first? Drop a comment below.
About the author: reventadata builds and publishes data extraction Actors on the Apify Store.
Top comments (0)