Used Car Deal Finder AI: mobile.de Qualified Car Deals, 10 Free

New underpriced mobile.de cars, ranked with risks flagged, in your inbox every week.

⚡ First 10 qualified car deals free · 💵 ≈ $0.022 per qualified car deal all in, the mobile.de scan on your account included — $0.006 of it is ours · 🤖 $0.03 per car with AI + tokens, on plans with model access or your key · ⏱️ demo in seconds

Try it in 30 seconds. Click Try it — the form is pre-filled with a free demo of 12 sample cars. Cars that fail your filters cost nothing here. → What a run costs · What the AI adds · Run it weekly

⚡ At a glance

What you get	A ranked sheet of mobile.de cars priced under comparable cars: % under market, margin estimate, risk flags, a first message to the seller
You provide	A mobile.de search link (or just a make like BMW) and, if you like, what you want to buy in plain words
Output	A ranked sheet (Excel, CSV, Google Sheets) plus an e-mail digest of what is new each run
Typical run	200 listings → ≈ 10 qualified car deals (1 in 20, estimate — not yet measured on a live search); the demo scores 12 sample cars in under a second
Cost of that run	≈ $0.23 rules only · ≈ $0.50 with the AI — ≈ $0.022 per car all in rules only, the mobile.de scan on your account included. Your first 10 qualified cars are free
Free tier	First 10 qualified car deals per account
Keys / setup	🔑 No API key where your Apify plan includes model access — otherwise your own Gemini / Groq key (a free key gives 1–3 AI verdicts per run, a paid key assesses every car)
Which plan	Free demo and your own list (ready rows): any plan · live search: a plan that runs Store Actors (else 0 cars, nothing charged)
Works with	Schedules and deal alerts every half hour, webhooks, Google Sheets, Slack, Microsoft Teams, Google Chat, Discord, Telegram, Make / n8n / Zapier, MCP & AI agents

🎯 What this Actor does

A used car deal finder for mobile.de — or any car site (AutoScout24, Kleinanzeigen) through rows you paste in list mode: it ranks every car against the comparable cars of the same search, drafts the seller message, and charges only for cars that pass your filters.

• Price against the market, not a guess. Each car vs the comparable cars of the same search (same make and model, ±2 years, similar mileage, same fuel): median, % under market, margin estimate. Under 30 cars loaded, the status warns the median is thin.
• Risks read for you. Accident, export-only, rental use, defects and missing HU, flagged from the listing and the German or English seller text — with the quoted sentence when the AI reads it.
• A message ready to send. One short first message in the listing's language, with one specific question.
• Only new cars, and price drops. A car you already received is never charged twice — unless its price fell 5 % or more since.
• Search from a make. No link? Type BMW, Audi, Mercedes-Benz, Volkswagen, Opel, Ford, Porsche, Škoda or SEAT and the search is built.

🚀 How to use it

1. Click Try it — the form is pre-filled with the free demo on 12 sample cars; it works as is.
2. Type a make ("Car make") or paste your mobile.de search link, and what you want to buy. Your search replaces the sample and runs live on your Apify account, capped by "Max source spend". Leave "AI assessment" on: your first 5 cars get live AI verdicts on any plan; all 10 free ones with Apify model access or a paid own key.
3. Press Start and open the dataset — qualified cars land best-first; the Outreach view holds the first message.
4. Schedule it weekly (Actions → Schedule) with your address in "E-mail me the new qualified car deals" — only new cars and price drops are delivered and charged.

📦 Already scraping mobile.de? Qualify your mobile.de scraper data

Pick dataset as the source and paste your last scraper run's dataset ID into Dataset ID — no second scrape, you pay only for the qualified cars. After every scrape: the chained-run recipe under Integrations.

📚 Learn more: Step-by-step tutorial — how to find underpriced cars on mobile.de with this Actor

🆚 Why this instead of running the scraper yourself?

	memo23/mobile-de-scraper · 101 users / 30 days	Used Car Deal Finder AI
Your time per 100 car deals	≈ 11 h screening 2,000 rows + ≈ 10 h vetting 100 cars (estimate: 20 s per row, 6 min per car)	≈ 0 h screening — ranked, reasoned, first message drafted
What you pay for	every row scraped: accident, export and overpriced cars	only qualified car deals — junk costs $0 here (the scan is billed on your account, capped)
Same 100 usable car deals, all in	≈ $1.61 for 2,000 raw rows (2,000 × $0.0008 + $0.005) — and you still sort them	100 ranked, explained cars: $2.21 rules only ($1.61 scan + 100 × $0.006) · $4.91 with the AI (+ 100 × $0.03 + 100 × ≈ $0.003 tokens)
Repeat runs	the same rows again next week	only new cars and price drops

vs other mobile.de scrapers and deal finders

Store listing read 2026-10-04	3x1t/mobile-de-scraper-ppr (150 users / 30 days)	blackfalcondata/mobile-de-scraper (115)	ivanvs/mobile-de-scraper (56)	This Actor
Your time per 100 car deals	≈ 21 h	≈ 21 h	≈ 21 h	≈ 0 h screening
What you get	raw rows; 43 of 50,491 runs failed	raw rows; tracks price changes	raw rows	each car scored, risks, message
100 usable car deals, all in	≈ $3.00 + sorting by hand	≈ $1.19 + sorting by hand	≈ $3.00 + sorting by hand	100 ranked cars, junk $0: $2.61 / $5.91
Price	$0.0015 per result + $0.002 start	$0.00059 per result + $0.005 start	$0.0015 per result	$0.006 / $0.03 per qualified car deal

Other used-car deal finders (read 2026-10-17)	What they cover · price per row	mobile.de?
swerve/il-car-deals	Israel (Yad2) deal scores · $0.005	no
ahmed_jasarevic/willhaben-scraper · unitbytes/offerup-scraper	Austria / US raw rows · $0.00085 / $0.0008	no
thomasthinks/kijiji-deal-finder · constructive_calm/gulf-used-car-scraper · outspoken_strategy/google-maps-lead-finder	Canada · the Gulf · Google Maps contacts	no

Free tool you already use: mobile.de saved searches and price rating badge — free and rated over the full market, so it wins there; it has no API, schedule, margin estimate or seller-text read.

💵 Pricing — what a run really costs

Pay per result: one event per qualified car deal, nothing else.

Event	When it is charged	Price
free-tier	your first 10 qualified car deals on this Actor, in any mode	$0.00
qualified-car-deal-basic	AI off — fields, market baseline, flags, rule score, message template	$0.006
qualified-car-deal-ai	AI on — plus the risk read with quotes, the score verdict, the reason, the seller message	$0.03

One free tier: your first 10 qualified car deals, whatever the tier. The mobile.de scan runs on your Apify account (≈ $0.0008 per listing plus $0.005 start, capped by "Max source spend"). Never charged by us: cars below the minimum score, failing a filter, without a market baseline, on your exclude list or already received.

How that compares — $0.006 is 1.4× the $0.00425 median of the Store search this card sits in (4× one raw mobile.de row, $0.0015), but one qualified car deal replaces ≈ 20 raw rows — $0.0003 per raw-row equivalent; the AI tier carries the premium.

Worked example — 200 listings → ≈ 10 qualified (estimate): scan $0.17 on your account + 10 × $0.03 + 10 × ≈ $0.003 tokens = ≈ $0.50 with the AI, ≈ $0.05 per qualified car deal; rules only ≈ $0.23, ≈ $0.023 per car.

At volume — a reseller, 5 searches × 300 listings weekly: ≈ $1.23 scan + ≈ 25 new cars × $0.033 → ≈ $2.06 a week, ≈ $9 a month (estimate).

AI tier, all in: ≈ $0.033 per qualified car deal ($0.03 here + ≈ $0.003 tokens via Apify or your own key) — 5.5× the basic price.

🤖 What the AI tier adds

AI off gives the price gap, flags, a rule score and a template message:

{"name": "BMW 320d Touring Aut. Navi Xenon, Privatverkauf", "ruleScore": 83, "score": 83, "label": "hot",
 "assessedBy": "rules", "flags": ["well_below_market", "price_rating_agrees"], "aiModel": null}

AI on (qualified-car-deal-ai, $0.03) — the same car, as the AI example run of 2026-10-17 wrote it with an own Gemini key (12 sample cars, aiModel verbatim):

{"name": "BMW 320d Touring Aut. Navi Xenon, Privatverkauf", "fitScore": 93, "score": 93, "ruleScore": 83,
 "label": "hot", "assessedBy": "ai",
 "dealReason": "22.1% under market median (€19,250), 99,500 km, 2018 model, very good price rating, 7 comparables.",
 "scoreReasons": ["22.1% under market median (€19,250)", "Clean history and accident-free (1st hand)", "Valid HU until 03/2027"],
 "riskFlags": [], "riskSeverity": "low", "messageLanguage": "de",
 "generatedText": "Hallo, ich habe Interesse an Ihrem BMW 320d Touring für 14.990 €. Ist das Fahrzeug noch verfügbar und kann ich kurzfristig einen Besichtigungstermin vereinbaren?",
 "recommendedNextStep": "Call the seller to confirm vehicle availability and arrange a prompt viewing.",
 "aiModel": "google:gemini-3.5-flash-lite"}

On the run's rental car it quoted the sentence behind its flag; the model moves the rule score by at most 15 points. Reading the German text and writing to the seller takes ≈ 6 minutes per car by hand (estimate based on reading 3 sample cars, 2026-10-04).

Try the AI on any plan: the demo shows labelled sample AI verdicts; your first real search gets 5 live AI verdicts on us — then a free Gemini key (aistudio.google.com/apikey) assesses a few cars per run, a paid Gemini / Groq key every car.

⚙️ Input

Flags for targetFlags (well_below_market, below_market, accident_hint, export_only and more) and every field of a row: API reference.

📤 Output

One dataset item per qualified car deal — export as JSON, CSV or Excel, or stream it to a webhook. Views: Shortlist, Outreach (contact route, seller, message), Market check, Risk check and Score breakdown.

Which score to use: sort by score; ruleScore is the rule part and fitScore the AI's — threshold and trend across runs on ruleScore; assessedBy says which scale score is on. The Score breakdown view has the parts. url is the field to map.

Upsert key: dedupeKey (id: + the mobile.de listing id) — one car is one row however many runs find it; itemId is a hash of it.

A DIGEST record ("vs last run: qualified 41 → 55, hot 5 → 9" and the top 10) and a client-ready REPORT page (OUTPUT.reportUrl) sit next to OUTPUT — the run record (RUN_SUMMARY its older name) with the funnel and outcome; a failed source sets sourceError, an unavailable AI aiError.

⭐ Found it useful? A review on the Store helps others find it.

🔁 Run it weekly

1. Open the saved task and click Schedule after your first real run (OUTPUT.scheduleTaskUrl), your address in notifyEmail: every run with new cars e-mails the digest with the trend vs the last run.
2. Leave "Never output the same qualified car deal twice" on: the next run charges only what is new — plus cars whose price fell 5 % or more. A second link keeps the memory; each client gets its own dedupeStoreName. A car charged under another search of the same memory comes as already-charged, never charged twice.
3. Paste a Slack, Teams, Google Chat or Discord channel webhook — or your Telegram bot's send link — into Send alerts to (sendAlertsTo, one per line, detected from the address): the digest lands in your channel; any other https URL gets the rows as JSON. For Google Sheets: the ready tracker sheet (dedupeKey first).
4. OUTPUT.history and the REPORT page show every run; a scheduled run scans twice the listings and says when it fills its cap; historyDatasetName keeps a running log; dedupeTtlDays re-delivers after N days (0 = never).

Weekly cost is the new cars only: 5 searches × 300 listings → ≈ 25 new cars a week after week 1 ≈ $2.06 a week with the AI, scan included (estimate until a scheduled run measures week 2).

⚡ Deal alerts every half hour

Schedule your task every 30 minutes with "Minimum score" 70, "Max listings to scan" 50 and Send alerts to: each run posts only new hot cars; a 50-listing scan costs ≈ $0.045 on your account.

🎯 Try it for your niche

Pick one in Start from a ready niche search and press Start (your own values win). Not yet measured live, so each caps its first run at 150 listings and $0.15.

Niche	What it looks for	Start
Diesel BMW estates under 25,000 EUR	BMW, diesel, 2017+, under 160,000 km, ≥ 10 % under market	preset bmw-diesel-estates
Private-seller bargains under 10,000 EUR	private sellers only, 2012+, ≥ 12 % under market	preset private-seller-bargains
Electric cars from dealers under 30,000 EUR	electric, dealers, 2020+, under 80,000 km	preset electric-dealer-stock

🔌 Integrations, automation and API

Run outcomes

• After every mobile.de scraper run: in the Integrations tab of memo23/mobile-de-scraper (or the scraper you use) add "Run succeeded → start this Actor" with sourceMode: dataset, datasetId = the finished run's resource.defaultDatasetId and enableAi: false — every scrape you already pay for is qualified automatically, rules only until you switch AI on.
• Google Sheets / Slack — one click from the Integrations tab.
• AI agents / MCP — every input is a plain typed field. Free first call: {} runs the demo, and {"sourceMode": "list", "itemsList": [...]} scores listings you already have; the snippets below run the real search, capped by maxDiscoveryChargeUsd.
• Webhook and CRM — any https URL in Send alerts to gets every car as JSON; for HubSpot / Pipedrive use a Make / Zapier step upserting on dedupeKey.

Minimal agent input — the fields the job needs; every other input keeps its default. It turns the cross-run memory off, so the same question asked twice returns the top cars both times:

{"searchUrls": ["https://suchen.mobile.de/fahrzeuge/search.html?isSearchRequest=true&s=Car&vc=Car&ms=3500"], "goalDescription": "clean BMW diesel estates", "maxPrice": 25000, "maxQualified": 25, "maxDiscoveryChargeUsd": 0.5, "dedupeAcrossRuns": false}

Which call runs what: {} and the untouched form run the free demo (OUTPUT.mode demo); your link or make starts the capped live search; your itemsList runs list, a datasetId runs dataset.

List-mode rows

sourceMode: "list" runs on any Apify plan, no scraper run: export your saved cars from AutoScout24, Kleinanzeigen or a spreadsheet and map the columns to the keys of the ready template linked under At a glance.

👥 Who is it for?

You are…	You run it to…	Start with
A used-car reseller or flipper	find the 10 cars worth calling today, with price gap, risks and a message	the BMW diesel estates preset, then your own link
A small dealer buying from private sellers	spot private cars under comparable cars before others call	the private-seller bargains preset
An export trader (DE → NL / PL)	scan a make across Germany weekly, keep clean, under-market cars	a make like Audi + a maximum price

🧠 How the AI works

Grounded, not generative: the model reads one car at a time and every claim is checked against the row.

🔒 Data, compliance and limits

• Public listing data only — what mobile.de publishes in the ad. No login. Not affiliated with mobile.de. A private seller's name, phone and e-mail are dropped unless you switch it on (GDPR).
• Limits: mobile.de shows at most 2,000 results per link — split a big market into several links. EUR prices; Germany's mobile.de in this version. A median from too few cars is never shown as a market price. Keep your use within mobile.de's terms and applicable law (GDPR, UWG).

❓ FAQ

How much will one run cost me? The default 200-listing scan costs ≈ $0.17 on your Apify account (capped by "Max source spend", $0.50 in the form); each qualified car deal is $0.006 (rules) or $0.03 (AI) here, the first 10 free — ≈ $0.23 rules only, ≈ $0.50 with the AI for ≈ 10 cars (estimate).

How long does a run take? The demo 0.1 s (measured); the rest are estimates dated 2026-10-18:

Do rule-scored cars use my free ones? Yes, while the AI is unavailable — except the last 5, the AI reserve, kept so your first AI-assessed cars are free too.

How do I send the messages safely? By hand, a few a day from your own mobile.de account; no links or off-platform contact IDs in the first message. The AI's message leaves links out and never promises a price.

More questions, presets and a Make / Zapier recipe: API reference.

🧩 More Actors from the same developer

Same pay-per-qualified-result model, closest to this Actor's buyers first:

• eBay Sold Listings Comps AI — Price the parts, wheels and accessories of a flip before you buy: clean eBay sold comps and a max buy price.
• Facebook Ad Library Winning Ads Finder — Your competitors' longest-running Facebook ads — hook, offer and angle broken down, one row each.
• Website Change Monitor — Every real change on your competitors' pages, explained, in your inbox every week.
• Amazon Bestsellers Scraper + AI — New Amazon products worth sourcing — rising, beatable, never repeated — in your inbox every week.
• 1688 Supplier Finder — New trusted 1688 factories for your product — costed and translated, in your inbox every week.

🆘 Support

Something missing or wrong? Open an issue on the Actor's page.

📝 Changelog

• 0.1.5 (2026-10-17) — One "Send alerts to" box for Slack, Teams, Google Chat, Discord, Telegram or any webhook — paste the link, the new deals land there; the form is shorter; A car you already paid for under another search comes through marked "already-charged" and is never billed twice; within hot / warm the AI-checked cars come first, and a "Score breakdown" view shows the rule and AI parts; Your own free Gemini key now works without typing a model name, the AI example on the page comes from a real run, and the ready niche searches cap their first run at $0.15.
• 0.1.4 (2026-10-04) — The title names mobile.de; the form says up front which plans the live search runs on; minimal example inputs; every output field named; a shorter page.
• 0.1.1–0.1.2 (2026-10-04) — first release: cars scored against comparable cars, risk flags, seller message, price-drop re-delivery, three ready searches, search from a make.