The comps are
on the card.
Ninja air fryer, $17.98 on the Mueller endcap. Sold comps $62–74, median $68.00. After the 13.25%, the promoted rate, the shipping shortfall and the box you clear $34.00 each — and there are four on the shelf. That’s $136.00 on one endcap, decided in four seconds, standing in the aisle.
Ninja AF101 air fryer, 4qt
86%Target · Mueller · aisle D14, endcap
$62–$74
$68.00
−$14.54
$19.46
$34.00
175%
6 days
“Is this worth the drive from S Lamar?”
6.1 miles round trip · $136.00 expected · $22.30 a mile. Yes. S Lamar also has 2 — I haven’t checked those since Tuesday.
Ninja AF101 ×4 pulled off the Mueller endcap — $136.00 expected
flipcity · confirmed 4m ago
Thursday 6:00am. Whatever is still on D14 goes to the 90% cut or goes back to the warehouse — the community has watched this store do it 11 weeks running.
Every fee, on the line
where it actually happens.
Discount is a vanity number. A 99% badge on something that sells twice a year is worse than 86% off a thing that moves in six days. So the card shows the receipt: landed cost in, comp out, every deduction named, and one net you can bank.
Ninja AF101 air fryer, 4qt
Target · Mueller · aisle D14, endcap · 4 on hand
scanned 14m ago by aislewalker · 97% accurate
$62–$74
$68.00
42 sold in 30 days
6 days
69% sell-through · 42 sold against 61 active
Target clearance, was $129.00
8.25% Travis County
what you actually paid
median of 42 sold, last 30 days
13.25% of total + $0.40
2.0% at the category's suggested rate
charged $12.95, label cost $15.62
14×12×12, 9.2 lb dim-weighted
$34.00
175%
50.0%
The fee line is your fee line
Store tier, category rate, promoted rate, ship-from ZIP, tax county — set once. Defaults are eBay Basic at 13.25% plus $0.40 and 2.0% promoted, which is what most people are actually paying.
Shipping is dim weight, not vibes
A 14×12×12 box at 9.2 lb bills as 12 lb from 78723 to zone 4. You quoted $12.95; the label is $15.62. That −$2.67 is 7.9% of the net and it never shows up in anyone's screenshot.
Comps are solds, and the sample is printed
42 sold in 30 days, against 61 active. Below 8 solds in 90 days we print the count and refuse to print a median — a median of three is a rumor with a dollar sign on it.
Velocity sits next to margin
Median 6 days to first sale. The same $34.00 held for 46 days is a completely different business, so the card never shows you one number without the other.
Then it multiplies by the shelf
4 on hand at aisle D14, scanned 14 minutes ago by a contributor at 97% accuracy. $136.00 sitting on one endcap, decided before you picked the first box up.
Three cards, three answers
Same shape, same math, three different answers. The buy price is the least interesting number on any of them.
Lego Icons Orchid 10311
75%Target · Mueller · aisle G03
$17.12
106%
4 days
Retired set, Q4 index 2.3×. Buy every one on the peg.
Pokémon 151 Elite Trainer Box
58%Costco · Southpark Meadows
$18.34
34%
3 days
Fast money, thin money. Fine at 2 units, not at 20.
Instant Pot Duo 6qt
62%Target · S Lamar · aisle D11
$0.26
1%
27 days
Comps are asks, not solds. 308 active listings is a wall.
Ranked by profit per mile, not by discount
goat looks at 11 stores inside your radius, prices what's flaggable at each one against your fee profile, weights it by that category's 30-day sell-through, then divides by the marginal miles. Stops under your floor get cut, and it tells you why.
4 stops
21.5 mi
$729
$33.90/mi
Target · Mueller
1.2 mi · 6 minAir fryer ×4, Lego Orchid ×3, Dyson filters ×2, Keurig ×2.
$186
Home Depot · Slaughter Ln
6.4 mi · 14 minRyobi kit at $0.01 confirmed by 96 people, bay 42. Bring the scanner.
$214
Dillard's · Barton Creek clearance center
9.1 mi · 17 minFrye boots sizes 7–10 intact, plus two Ugg returns on the 75% rack.
$268
Kohl's · Arboretum
4.8 mi · 11 minKeurig K-Mini ×4 at $24, and the 30% code stacks. Marginal but on the way home.
$61
Walmart · Ben White
11.3 mi · 22 minBelow your $12.00/mi floor. goat cut it — the same two units show at Ben White every week and never move.
$74
Marginal miles, not total miles — a store on the way home costs almost nothing to add, which is why Kohl’s survives at $12.71 and Walmart doesn’t at $6.55.
- Send you to a store whose last scan is four days old without saying so on the pin.
- Count units it hasn't seen — expected net uses on-hand counts, never a guess at what's in the back.
- Rank on discount. An endcap of 90%-off calendars ranks below three air fryers, every time.
- Hide the cut stops. They stay on the map, greyed, with the number that killed them.
Ask it like you’d ask a partner
goat runs the same tools you would: scan, store check, sold comps, price history, fee math, route. The difference is that it already knows your ship-from ZIP, your store tier, your $12.00 floor and the four stores you actually go to. It answers in dollars and store names, and when it hasn’t checked something it says so.
Plan me a sourcing route for Saturday morning, $12/mi floor
11 stores in range, 4 clear your floor. 21.5 miles, about 2h 40m, expected net $729 — that's $33.90 a mile. Start at Target Mueller: the air fryer is 4 deep on the D14 endcap and it was scanned 14 minutes ago.
Margin tells you if. Velocity tells you how many.
A number you can't sell is inventory. Every category on the card carries sold-against-active, a median days-to-sell, and a Q4 index — so you know in July whether to buy four or forty.
Buy in July, sell in November. The same air fryer that nets $34.00 in August nets $50.95 against the November median, because the median comp moves from $68.00 to $88.00 and the fees are a percentage.
Q4 is not one season
Small kitchen appliances peak in November at 2.1×. Retired Lego peaks at 2.3× but starts climbing in September. Patio goes to 0.4× and stays there until March — the same clearance rack is a buy and a trap depending on the aisle.
Saturation shows up first in active count
When 40 people read the same 86% badge, active listings move before sold prices do. The card shows both, so you can watch a comp get crowded in real time instead of finding out at week three.
Hold time is a cost you already pay
A 46-day median is a month and a half of capital you can't put on the next endcap. Days-to-sell sits beside the margin on every card, because the fast $17.12 turns over eleven times while the slow $182.44 turns over twice.
A ledger, not a highlight reel
Ten closed flips from one Austin seller across nine stores, with fees broken out and the two that went sideways left in. Goatavo writes the row for you when you mark something sold — buy, tax, shipping both ways, platform fees, promoted spend, net, days held.
| Item | Sourced at | Off | Landed | Sold | Fees | Net | ROI | Days |
|---|---|---|---|---|---|---|---|---|
Ninja AF101 air fryer, 4qt | Target · Mueller | 86% | $19.46 | $68.00 | −$14.54 | $34.00 | 175% | 6d |
Ryobi 18V drill/driver kit | Home Depot · Slaughter Ln | 100% | $0.01 | $84.00 | −$18.42 | $65.57 | — | 11d |
Lego Icons Orchid 10311 | Target · Mueller | 75% | $16.22 | $41.50 | −$8.16 | $17.12 | 106% | 4d |
Dyson V8 filter set ×3 | Walmart · Sunset Valley | 86% | $32.43 | $96.00 | −$19.88 | $43.69 | 135% | 9d |
Hampton Bay patio, 5-piece | Home Depot · Brodie Ln | 85% | $96.36 | $340.00 | −$61.20 | $182.44 | 189% | 22d |
Pokémon 151 ETB ×2 | Costco · Southpark Meadows | 58% | $108.23 | $178.00 | −$33.10 | $36.67 | 34% | 3d |
Frye Melissa boots, sz 8 | Dillard's · Barton Creek | 75% | $83.89 | $168.00 | −$32.44 | $51.67 | 62% | 31d |
Keurig K-Mini ×4 | Kohl's · Arboretum | 72% | $103.92 | $216.00 | −$41.90 | $70.18 | 68% | 14d |
Instant Pot Duo 6qtthin | Target · S Lamar | 62% | $38.94 | $61.00 | −$15.86 | $6.20 | 16% | 27d |
Vitamix E310, returned + relistedreturn | Costco · Cedar Park | 45% | $194.61 | $268.00 | −$78.24 | −$4.85 | −2% | 46d |
| 10 flips | 9 stores · one seller | $694.07 | $1,520.50 | −$323.74 | $502.69 | 72% | 12.5d median |
Net = sold − platform fees − landed cost. Landed cost includes sales tax and any shipping you paid to get it. The penny row has no meaningful ROI denominator, so it is counted in net only.
Two rows are here on purpose: the Instant Pot cleared $6.20 in 27 days and the Vitamix came back “not as described” and closed at −$4.85. A ledger that only shows wins is a screenshot, not a business.
The comps are the product. So they're the paid part.
We'd rather tell you that plainly than hide it behind a trial. The feed, the shelf counts, the aisle numbers and the price history are free forever — because that's the part everyone needs. The comps, the fee math and the route are Pro, because that's the part that makes you money.
Ninja AF101 air fryer, 4qt
86%Target · Mueller · aisle D14, endcap · 4 on hand · 14m ago
Nothing is blurred out of spite. The free card still tells you the price, the aisle, the count and the history — enough to decide whether to walk over. Pro tells you whether to buy all four.
Free, forever
no card, no trial clock- The whole feed — clearance, penny list, drops, coupons, cards
- Store-level shelf counts and aisle numbers, with a timestamp
- 90-day price history and the all-time-low flag
- 5 live store checks a day
- Alerts, batched every 15 minutes
- The community, the confirmations and the accuracy scores
Plenty of people run a whole side hustle on this tier and do the comps in a second app. We’d rather you didn’t, but the door isn’t locked.
Pro — the numbers
launch price, locked for life- Sold comps on the cardeBay, Mercari and StockX solds — 30 and 90 day, range and median, not asks
- Profit math with your fee profileYour store subscription rate, promoted rate, ship-from ZIP and tax county
- Sell-through and days-to-sellSold against active, by category and by month, so you can see saturation
- The route plannerStops ranked by expected profit per mile, with your own floor
- Unlimited live store checksInstead of 5 a day — check the whole route before you leave the house
- Instant alertsNot batched. Fifteen minutes is the difference between four units and none
- Ledger exportCSV of every flip with fees broken out, ready for a Schedule C
One flip like the air fryer pays for a year of it. We’ll show you that in the ledger — net earned against what you paid us — and if it’s negative the ledger says so in the same place.
No dark pattern here: the free tier is not a demo, the locked rows say exactly what they contain, and nothing in the app ever covers a price, an aisle or a stock count with a paywall.
Most clearance is not a flip
Roughly one in nine flagged items clears a $15 net after fees, shipping and a realistic comp. We show you the other eight anyway, with the reason they failed, because knowing why something is a bad flip is most of the skill.
Asks are not solds
An item listed at $74 by 308 people is not worth $74. Goatavo only counts completed sales, shows you the active count next to it, and refuses to show a median under 8 solds in 90 days.
Sell-through is the real gate
A patio set at 22% sell-through and 46 median days is a garage, not a business. The card puts velocity next to the margin so you can't read one without the other.
Fees drift, and they drift up
Final value rate, category rate, promoted rate, shipping under-collection — four lines that only move one way. Change one in your profile and every open card reprices, including the ones already sitting in your saved list.
Returns eat the blended number
Small appliances run about 4% returns. Two in fifty, each one eating the unit, the box and both labels at $51.80, turns a $34.00 net into $30.57 blended — and one of them comes back missing the basket.
Gates, brands and quantity limits
Amazon gating, brand restrictions, one-per-customer at the register, and clearance that never restocks. The card flags a gated category before you drive, not after you buy eleven.
Not every deal is a flip
Most of what Goatavo finds is worth buying because you wanted it, not because you can resell it. The floor is a setting: mine sits at $15 net, brickvault runs $40 and buys a third as often, and the feed re-sorts either way.
Any tool that finds shelves faster makes those shelves emptier. That is a real cost to you and we’re not going to pretend otherwise. What we do about it: alerts are per-store and per-radius rather than national, the feed ranks by net instead of by discount so the viral 99% badges stay where they belong, and velocity is on every card so you can watch a comp getting crowded before your capital is in it.
Full-time, part-time, and one guy with an endcap habit
Beta figures, self-reported and tied to a ledger. The accuracy score beside each handle is their community confirmation rate, which is the only number here we can vouch for.
“Eleven to twelve pulls a scan-through, and I haven't typed a model number into a comp app since March. The net per unit is there when the barcode clears.”
“It cut a Walmart stop I'd been making for a year. Eleven miles for $74 that never sold. I hadn't done that division once.”
“The shipping shortfall line is the one that changed how I price. I was quoting $12.95 on a nine-pound box and calling it free shipping.”
“Q4 index of 2.3 on retired sets, right on the card in July. I bought 40 Orchids in the summer and sold them into November.”
“Bay 42, confirmed by 96 people, twelve minutes old. That's a different sentence than 'someone said Home Depot'.”
“It told me to skip the Instant Pot. 308 active listings against 96 sold. I'd have bought six of them a year ago.”
Four seconds instead of ninety
Everyone doing this already has a comp app. The problem isn't that it's wrong — it's that it costs you a minute and a half per item while you stand in an aisle with eleven things to check and a store that closes at nine.
| A second app in the aisle | Goatavo | |
|---|---|---|
| Where the comps live | A second app you open in the aisle | On the card, before you pick it up |
| Time per decision | 70–90 seconds of typing a model number | Under 4 seconds — scan, read net, move |
| Fees in the number | You do 13.25% in your head | Your fee profile, applied per category |
| Shipping | Guessed, then eaten | Dim weight from the box you actually use |
| Shelf count | Not available | On-hand, aisle, and who last scanned it |
| The rest of the store | Only what you thought to check | Everything flagged, ranked by net |
| The drive | Whatever order the stores are in | Ranked by expected profit per mile |
Before you join
If something here isn’t answered, ask — replies come from the people building it, not a help desk.
Where do the sold comps come from?+
Completed sales on eBay, Mercari and StockX, pulled per item and refreshed daily. We show the range, the median, the count and the active-listing count beside it. If there are fewer than 8 solds in 90 days we show the count and refuse to print a median at all.
Does the profit math use my fees or generic ones?+
Yours. You set your store subscription tier, your category's final value rate, your promoted-listing rate, your ship-from ZIP and your tax county once. Every card after that shows a net you can actually bank, and changing a rate reprices the cards you already saved. Before you set anything it assumes eBay Basic.
Is the shelf count real or crowd-sourced?+
Both, labeled. Where a retailer exposes on-hand counts we read them directly. Everywhere else it's a community scan with a timestamp, a confirmation count and the scanner's accuracy score. The card never says '4 in stock' without saying who said so and when.
How is the route planner ranked?+
Expected net divided by marginal miles, against a floor you set. Expected net is per-unit net times units on hand times that category's 30-day sell-through, so a stop with eleven slow units doesn't beat a stop with three fast ones. Stops under your floor get cut with the reason shown.
What's free and what's Pro?+
The feed, the shelf counts, the aisle numbers, price history, the community and 5 live store checks a day are free forever. Comps, the profit math, sell-through, the route planner, unlimited checks, instant alerts and ledger export are Pro. Founding members lock the launch price for life.
Will this flood the deals I source?+
Honest answer: yes, at the margin. Two things blunt it. Your alert radius is yours and defaults to 15 miles, so a Mueller endcap doesn't get broadcast to Dallas. And there is no tier above Pro — nobody can pay to get the same alert ahead of you, because Pro is already the fastest lane and it costs everyone the same.
Can I export for taxes?+
Yes. CSV with buy, tax, shipping in and out, platform fees, promoted spend, net and days held, per item. It maps to a Schedule C without a spreadsheet evening. Mileage from the route planner exports too.
Do I have to be a full-time reseller?+
No, and most people here aren't. Two evenings a week and one Saturday route is the median. The profit math is a toggle — turn it off and it's a clearance app.
Three surfaces feed the same profit math
Clearance, the penny list and drops all land in one feed, on one card shape, with the same net number attached. Sourcing is the input; the ledger is the output.
Stop opening a second app.
The math is on the card.
Turn the profit math off and Goatavo is a clearance app. Turn it on and it's the only place the comps, your fee profile, the shelf count and the drive land in one number. Tell us your metro and we'll seed its solds before the rest of the map.
- ✓A launch date before it's public
- ✓Founding Pro pricing, locked for life
- ✓A vote on the next marketplace we index