By the end of this tutorial you will have a short Python script that pulls adverts from willhaben.at, the Austrian classifieds site, into structured rows: asking price, seller type, location, photo links and the advert's own address on the site.

Willhaben Scraper (reventadata/willhaben-scraper) is the Apify Actor built for this site. You give it one or more keywords, it walks the public search pages, and it returns one row per advert.

This guide runs it from Python with the Apify Python client. The sample rows below are copied from a real run of the Actor's example task on 18 November 2026, not from an invented example.

Prerequisites

You need three things:

  1. Python 3.11 or newer, because the Apify client for Python 3.x requires it.
  2. A free Apify account. You can sign up with an email address or a GitHub login.
  3. Your API token, which belongs to your Apify account.

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 the public pages, so you do not need a login on willhaben.at.

What you'll achieve

You will search willhaben.at for the keyword iphone, read the rows in Python and print one line per advert: the price as the site shows it, the town, and whether the seller is private or a business.

The same input works for a buyer comparing asking prices, a seller checking what similar adverts ask, or an analyst building a table by Austrian federal state. Only the input changes. Every section of willhaben.at returns the same columns, so the script stays the same when you switch to real estate or cars.

Step 1: Open Willhaben Scraper

Open the Willhaben Scraper page on Apify Store. The Try for free button there opens the Actor in Apify Console. This guide uses the Python client, so you only need the Actor name, reventadata/willhaben-scraper.

Step 2: Choose the search

The search is described by a few input fields:

  • vertical: the section of willhaben.at. marketplace (Marktplatz, the general classifieds and the default), realEstate (Immobilien) or cars (Gebrauchtwagen).
  • searchQueries: keywords, one per line. Marktplatz only. Each keyword is searched separately, and an advert that two keywords both find is pushed once.
  • searchPath: an optional category or region page, copied from the address bar of willhaben.at. It must start with the section's own path, for example immobilien/mietwohnungen/wien. A path that carries a ? filter fails the run, and the run's message says to put keywords in searchQueries instead.
  • maxItems: a hard cap on rows across all keywords. The schema default is 100.
  • maxPagesPerSearch: an optional cap on result pages for each keyword. Each page holds 100 adverts, and the site stops at 999 pages.

The example task of this run sends exactly this input:

{ "searchQueries": ["iphone"], "vertical": "marketplace", "maxItems": 10, "maxPagesPerSearch": 1 } 

Read it as: one keyword, the marketplace section, at most ten rows, and at most one results page for that keyword.

Step 3: Run it from Python

This script starts the Actor, waits for it, reads the dataset and prints one line per advert:

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: the same input as the example task run_input = { "searchQueries": ["iphone"], "vertical": "marketplace", "maxItems": 10, "maxPagesPerSearch": 1, } # 3. Start the Actor and wait for it to finish run = client.actor("reventadata/willhaben-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)} adverts.\n") # 5. Print one line per advert: price as shown, town, seller type, link for item in items: seller = {True: "private", False: "business"}.get(item.get("isPrivateSeller"), "not stated") print(f"{item.get('priceDisplay')} {item.get('location')} {seller}") print(f" {item.get('url')}") 

The output starts like this. The first two of the ten lines are shown:

Retrieved 10 adverts. € 13 Oberhofen im Inntal private https://www.willhaben.at/iad/kaufen-und-verkaufen/d/griffin-kfz-ladegeraet-fuer-iphone-apple-auto-charger-lightning-connection-iphone-1511305679/ € 700 Theresienfeld private https://www.willhaben.at/iad/kaufen-und-verkaufen/d/iphone-16-pro-max-2069177942/ 

Understanding your results

The rows have one fixed set of columns. The ones you will use most:

  • price (a number) and priceDisplay (the text as the site shows it, for example "€ 1.389"). currency is EUR when the displayed price starts with the euro sign.
  • oldPrice and priceReducedAt: the previous price and the date when the seller lowered it. Three of the ten rows carry both.
  • isPrivateSeller: true for a private seller, false for a business.
  • sellerName: the business name. Six rows have it, and the four private rows have it empty.
  • location, postcode, state, district, latitude and longitude. All ten rows have coordinates.
  • title, description, url and publishedAt. The description is the start of the advert text as the search list shows it, ending in ... when the site shortens it. In this run seven of the ten descriptions are shortened, at 252 to 257 characters. The other three run 16 to 38 characters and are not shortened.
  • imageUrl, imageUrls and imageCount. The rows hold one to five photos.
  • isBumped: true on six of the ten rows.

One row from the same run. The imageUrls list is cut to its first entry, and the searchUrl field and the fields that are empty on this row are left out:

{ "id": "1543751976", "url": "https://www.willhaben.at/iad/kaufen-und-verkaufen/d/apple-iphone-15-pro-128-gb-titan-blau-1543751976/", "adType": 67, "title": "APPLE iPhone 15 Pro 128 GB Titan Blau", "price": 380, "currency": "EUR", "priceDisplay": "€ 380", "description": "Verkaufe mein Apple iPhone 15 Pro mit 128 GB Speicher. Das iPhone befindet sich optisch in einem sehr guten Zustand und wurde stets sorgfältig behandelt. Details: Modell: iPhone 15 Pro Speicher: 128 GB Akkukapazität: 81 % Optischer Zustand: Sehr gut Voll...", "location": "Wien, 14. Bezirk, Penzing", "postcode": "1140", "state": "Wien", "district": "Wien", "country": "Österreich", "latitude": 48.201813, "longitude": 16.275673, "publishedAt": "2026-11-18T08:30:58Z", "changedAt": "2026-11-18T08:30:58Z", "isPrivateSeller": true, "sellerId": "26801749", "imageUrl": "https://cache.willhaben.at/mmo/6/154/375/1976_114157453_hoved.jpg", "imageUrls": [ "https://cache.willhaben.at/mmo/6/154/375/1976_114157453.jpg" ], "imageCount": 5, "isBumped": false, "categoryIds": [ 2691, 2722, 2724, 5013241 ], "vertical": "marketplace", "searchQuery": "iphone" } 

The search list does not carry a phone number or an e-mail address. The Actor does not open the advert pages, so the full advert text is not in the rows.

Pro tips

  1. Keep keywords apart. Each line in searchQueries is its own search. Put a second product on a second line, and the rows carry searchQuery so you can split them afterwards.
  2. Check the id, not the title. In this run two rows share a title, a price and a location, and each has its own id. The Actor keeps one row per id.
  3. Cap the pages first. maxPagesPerSearch: 1 gives you a quick sample of one results page before you set a bigger maxItems.
  4. Use searchPath for a region or category. Copy the path after willhaben.at/iad/ from the address bar. realEstate and cars take only a path, not keywords.
  5. Compute your own totals. Sum price in Python, as you would for any list of numbers. Use priceDisplay only for showing the price, because it is text with a euro sign and a dot for thousands.

Troubleshooting

  • The run FAILED with a message about the input. A searchPath that does not start with its section's path, or one with a ? filter, stops the run before any search. The message says what to fix.
  • The run SUCCEEDED with 0 rows. The site answered and nothing matched. The status message says so. Try a shorter keyword.
  • Fewer rows than maxItems. The search had fewer adverts, or maxPagesPerSearch stopped it first. The log ends with the number of rows pushed.
  • Read the log. The run's log lists what was skipped and why, and ends with the number of rows pushed.

willhaben.at's terms of use and the GDPR apply to what you do with the rows. Check them before you store or publish personal data about sellers.

Next steps

For current pricing and details, see the Store page.

To repeat a search, save it as an Apify task and attach a schedule, which runs an Actor or task on a cron expression. Datasets export as JSON, CSV, Excel and more (dataset docs).

What are you planning to search for on willhaben.at? Drop a comment!

About the author: reventadata builds and publishes data extraction Actors on the Apify Store.