Ask ten resellers how they decide what to pay for something at an estate sale or a thrift store, and most will describe a feeling: "that looks like a $40 item, I'll go to $12." The feeling is wrong in both directions — too cautious on stuff that turns over fast, way too aggressive on stuff that just looks valuable. There's a formula underneath the feeling, and once you've run it a few times it takes less time than picking the item up and turning it over.
Work backward from the comp, not the price tag
The number that matters isn't what the item is worth. It's what it actually sold for, recently, in the condition yours is in. Two ways to check, and they're not the same tool:
- eBay's plain "Sold Listings" filter (the checkbox on a normal search)
- only shows about the last 90 days, mixed in with active listings unless you
- filter carefully.
- Product Research, inside Seller Hub under the Research tab — the tool
- reseller old-timers still call by its former name, Terapeak — is free for
- every eBay seller and pulls sold data back two to three years, filterable by
- category, condition, and even buyer location. It's the more reliable read on
- a real going rate, especially for anything with any seasonality to it.
Whichever you use, pull 3–5 comparable sold listings, not active ones (active listings are asking prices, not proof anyone paid them), and use the median, not the highest one you found.
The formula
Max buy price = (expected sale price) − (platform fees on that sale) − (shipping you'll pay) − (packaging) − (your minimum acceptable profit)
Everything on the right side except the last term is close to fixed once you know the sale price. The last term is a number you choose ahead of time — $10, $15, whatever makes an hour of sourcing, listing, and packing worth it to you — and it's the part gut-feel pricing skips entirely.
A worked example
Comps say a jacket like the one on the rack sells for about $42 total (item price plus the free shipping you'd absorb into it). Say it costs you about $7 to ship and $0.75 in mailer and tape, and you want at least $15 profit before you'll buy it.
- Final value fee: 13.6% × $42 = $5.71
- Per-order fee (orders over $10): $0.40
- Shipping label: $7.00
- Packaging: $0.75
- Total costs before cost of goods: $13.86
- Net proceeds if the item cost you $0: $28.14
- Minus your $15 profit floor: max buy price = $13.14
Round down to $13. If the tag says $18, walk away — not because $18 is a bad price in the abstract, it's just above what this specific item, at this specific comp, can carry and still pay you $15.
Build in a discount for stuff that sits
The formula above assumes the item sells promptly. It won't always. An item that takes four months to sell ties up the $13 you spent on it for four months, and if it needs a price cut to finally move, your realized profit is lower than the formula promised. Two adjustments that account for this without turning the math into a spreadsheet project:
- Slow or trend-dependent categories (fast fashion, dated electronics,
- anything seasonal you're buying off-season): shave 15–25% off your max buy
- price as a liquidity discount.
- Fast, evergreen categories (name-brand basics, tools, small
- electronics with steady demand): the formula as written is usually fine
- without an adjustment.
The 30% gut-check, for when you don't have five minutes
Standing in a thrift store, you're not going to run full fee math on every item. A rough version of the same formula, useful for a first pass: don't pay more than about 30% of the expected sale comp for anything in a normal-turn category, and closer to 20% for anything slow or seasonal. It's less precise than the full formula — it bakes in a rough profit floor rather than a specific dollar one — but it's built on the same logic and it's fast enough to use with a phone in one hand.
What this means at checkout
- Check a sold comp before you commit, not a listed price. Product
- Research if you have Seller Hub access; the 90-day sold filter if you don't.
- Subtract fees and shipping before you subtract profit, not after —
- profit is what's left, not a percentage you tack onto the sticker price.
- Set your profit floor before you shop, so it isn't a number you talk
- yourself out of standing at the register.
- Discount further on anything slow-moving. A great margin on paper means
- nothing if the item takes half a year to realize it.
PalmFlow won't tell you what to pay before you buy — that's still your judgment call. But it logs the real cost basis, fees, and sale price on every item once you do, so you can check whether your ceiling actually held up after the fact, item by item, not just on the ones you remember.