Used Car Deal Finder AI: mobile.de Qualified Car Deals, 10 Free on the Apify Store →

How to find underpriced cars on mobile.de with Used Car Deal Finder AI

Tutorial article (≈ 900 words) for the Apify blog guest program, dev.to, Medium, Hashnode. Drafted with the listing; the owner publishes it and records the URL in brief.json → marketing.links.

The problem

A used-car reseller opens mobile.de, searches "BMW 320d Touring, diesel, up to 25,000 EUR" and gets 2,000 cars. Which of them are priced under what comparable cars sell for, clean, and worth a phone call today? By hand it is the same loop for every car: open the ad, read the German seller text for "Unfallschaden", "Export" or "Bastlerfahrzeug", look for comparable cars with similar mileage and year, do the median in your head, and write the first message. About 20 seconds a row to screen and about 6 minutes a car to vet (an estimate from reading sample cars) — more than 20 hours for 100 cars. The raw mobile.de scrapers on the Apify Store stop at the 2,000 rows: they charge every row, accident cars and overpriced cars included.

Step by step

Click Try it. The form is pre-filled with a free demo on 12 sample cars. Nothing is charged. In under a second you get the shortlist: 3 qualified cars out of 12 — the others were filtered for free (one accident car, six not far enough under market, two without a market baseline).
Read the shortlist. Each row has the price, how far it is under the median of comparable cars (the best sample car: 22.1 % under a €19,250 median of 7 cars, a margin estimate of €4,260), a score, a hot / warm / cold label, risk flags and a one-line reason. The Outreach view holds the first message to the seller in German.
Paste your own search. Build the search on mobile.de, copy the link into "Your mobile.de search link" (or type a make like BMW), set "Max price", "Min year" and "Max mileage" and press Start. The Actor runs memo23/mobile-de-scraper on your Apify account, capped by "Max source spend" ($0.50 by default).
Let the AI read the text. With "AI assessment" on, the model reads the seller text and flags accident hints, export-only sales, rental use, defects and a missing HU — each flag needs a quoted sentence from the ad — and writes the message with one question for the seller.
Export or forward. The dataset exports as CSV or Excel; the REPORT page is a client-ready shortlist.

What a run costs

You pay for the scan on your own account and one event per qualified car deal here. A 200-listing scan is about $0.17 ($0.0008 a listing plus a $0.005 start). Ten qualified cars cost $0.10 with rules only and $0.40 with the AI plus about $0.03 in tokens — and the first 10 qualified cars are free, so the first search costs only the scan and tokens. The second run of the same search delivers and charges only new cars, plus any car whose price fell 5 % or more.

Automate it

Open the task the first real run saves, click Schedule and put your address in the e-mail field: every week the digest lists the new cars with price, gap under market and a link, plus the trend against last week. For a pipeline, the API call is a few lines:

from apify_client import ApifyClient

client = ApifyClient("<YOUR_API_TOKEN>")
run = client.actor("<your-apify-username>/used-car-deal-finder-ai").call(run_input={
    "sourceMode": "actor",
    "searchUrls": ["https://suchen.mobile.de/fahrzeuge/search.html?isSearchRequest=true&s=Car&vc=Car&ms=3500&fuels=DIESEL"],
    "maxPrice": 25000, "maxQualified": 25, "maxDiscoveryChargeUsd": 0.5,
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["name"], item["priceEur"], item["underMarketPct"], item["score"], item["url"])

The tool is unofficial and reads only what mobile.de publishes in the ad. A private seller's phone and name are dropped unless you switch it on.

Used Car Deal Finder AI: mobile.de Qualified Car Deals, 10 Free on the Apify Store →