Building a buy box for Kleinanzeigen: how to research what an item really sells for
Every guide to flipping on Kleinanzeigen — including mine — obsesses over speed: be first to message, be faster than the native alert, beat the other five buyers. That's all true, and all worthless without one thing sitting underneath it. If you're first to a bad deal, congratulations: you've just overpaid faster than anyone else.
The thing underneath speed is a number. Before you chase a listing, you should already know the most you're willing to pay for that item — worked out calmly, in advance, from what it actually resells for. Resellers call that a buy box: a pre-decided maximum buy price for the things you hunt. This is how you build one, and how to research real sold prices in a market that mostly hides them.
Asking price is not sold price (the trap that eats beginners)
Here's the single most expensive mistake in used-goods arbitrage: pricing your buy off what things are listed for instead of what they sell for. They are not the same number, and the gap is where your margin dies.
On Kleinanzeigen especially, the asking price is soft. Most private listings are VB (Verhandlungsbasis, open to offers), which means the number in the title is an opening position, not a transaction. A stack of identical items sitting at "120 € VB" for three weeks tells you what sellers hope to get, not what a buyer actually paid. If you build your buy box off other people's asking prices, you inherit their optimism — and you'll consistently overpay, then wonder why the flip barely cleared costs.
So the whole game of comps research is getting from asking prices (easy to see, misleading) to sold prices (harder to see, the truth).
Where to actually find sold prices
Kleinanzeigen itself doesn't hand you a clean sold-price history — a listing just disappears when it's gone, and you rarely learn what it went for. So you triangulate from a few sources:
- eBay "sold / completed" listings — your gold standard. This is the most valuable free comps tool most flippers underuse. On eBay, filter search results to Verkauft / Sold and you see the actual prices completed transactions closed at, with dates. That's real money changing hands, not wishful asks. For anything with a liquid eBay market — GPUs, consoles, sealed TCG, brand electronics — this is the closest thing to ground truth you'll get, and it's what you should anchor your buy box to. Look at the range and the recency, not a single lucky high sale.
- Current Kleinanzeigen asks — for context, adjusted down. Still useful: scan what the same item is listed at right now to gauge supply and the ceiling. Then mentally discount. A rough rule of thumb many sellers use is that used goods change hands somewhere around half to two-thirds of new retail depending on category and condition — but treat that as a sanity check, not a target. The real number comes from sold data, not a percentage.
- Category-specific references. Sneakers have StockX and similar resale indices with real transaction prices. Cars have Kleinanzeigen's own valuation tool and mobile.de asking data. TCG has set-level price trackers. Where a specialist price source exists for your niche, it usually beats generic guessing.
- Your own closed flips. The best comp is what you actually sold the last one for, through your channel, after your costs. Keep a running log. After a dozen flips in a category you'll know the number cold, and you'll need eBay less.
The discipline is the same across all of them: prefer prices that reflect a completed sale over prices that reflect a hope.
Turn comps into a buy box
Once you know the realistic resale, the max-buy falls out of simple subtraction. Don't price to "some profit" — price backwards from your net.
Max buy = expected resale price
− selling-platform fees
− shipping & packaging (both directions if returns happen)
− a risk buffer (condition surprises, slow sale, dead stock)
− your target margin
Worked example. Say eBay sold data shows a sealed item you hunt reliably closes around 90 €:
- Selling fees + payment processing: roughly 10–12 % → call it −10 €
- Shipping + packaging to the buyer: −6 €
- Risk buffer (10 % for the ones that arrive not-as-described, sit unsold, or need a partial refund): −9 €
- Your target margin (why you're doing this at all): −25 €
That leaves a max buy of about 40 €. So when a monitor pings you with that item at "35 € VB," you already know it's inside the box and you move now — no maths, no second-guessing, no talking yourself into a bad number at 11 p.m. And when it's listed at "70 € VB," you know instantly it's not a deal no matter how fast you saw it, and you don't waste a message or a pickup on it.
That's the real payoff of a buy box: it converts a fast alert into a fast decision.
The costs flippers forget (and why the buffer is non-negotiable)
The reason beginners' "profit" evaporates is almost always uncounted costs. Build them into the box so they can't surprise you:
- Versand both ways. Shipping to your buyer is obvious; the return you eat when something's "not as described" is the one people forget — plus packaging and label costs, small but real per unit.
- Platform fees. eBay and other resale channels take a cut plus payment processing. Know your actual percentage.
- Abholung logistics. If the Kleinanzeigen deal is pickup-only, factor the fuel and the hour. A 20 € margin an hour's drive away isn't a 20 € margin.
- Dead stock and the dud rate. Money tied up in something that sits for two months has a cost even if it sells at your number; and across enough flips some fraction go wrong — damaged, fake, ghosted, refunded. The risk buffer is you pricing for the average outcome, not the best case.
A buy box that ignores these isn't a buy box, it's a wish. The subtraction is the whole point.
Build the box before you need it
The reason to do this research now, cold, is that the moment you'll use it is the worst moment to think. A great listing appears, your heart rate goes up, four other buyers are presumably typing — that is not when to be estimating resale value and doing fee maths in your head. You want to glance at the price, compare it to a number you already trust, and either fire off your message or move on. So the workflow that wins is: research the category → set a max-buy → then turn on the alerts. The alert tells you fast; the buy box tells you whether to act. Neither works without the other.
Where a monitor fits
I flip on Kleinanzeigen and I built a small tool for the "fast" half — so let me be clear about which half it is. It's an AI agent that watches one eBay Kleinanzeigen search and pings my Discord when a new matching listing appears (about once a minute paid, every 15 minutes free), filtering each listing's text — title, description, price — against rules I write, via Google's Gemini with my own key. Kleinanzeigen-only, Discord-only, text only (no image analysis).
What it deliberately does not do is tell you what to pay. It has no idea what an item resells for, and shouldn't pretend to — the buy box is your homework, and it's the half that actually protects your money. A tool can get you to the listing first. Only your research decides whether being first was worth it.
The takeaway
Speed without a number is just a faster way to overpay. Do the comps work: anchor to real sold prices (eBay's completed listings are the best free source), discount other people's optimistic asks, and subtract every cost — fees, shipping both ways, a risk buffer, your margin — to land on a max-buy you decided in advance. Then, and only then, let the alerts run. When the ping comes, you won't be guessing; you'll be executing a number you already trust.
The monitor mentioned above is Schnappalarm — start free. It watches one Kleinanzeigen search and pings your Discord on new matching listings, filtered by rules you write. It handles the "first to see it" half; the buy box is yours. Kleinanzeigen-only, Discord-only, bring your own Gemini key.