| ● |
PB345 |
Light-up rubber duck, 3-pack |
2,840 |
96 |
3.4% |
4.1% |
steady |
no longer needed |
$1,248 |
$2,410 |
$25.10 |
38% |
normal for the category |
| ● |
PB106 |
Slant board, adjustable |
410 |
58 |
14.1% |
5.2% |
rising |
not as described |
$2,842 |
$5,180 |
$89.30 |
72% |
the listing over-promises |
|
A return rate is a listing metric before it is a quality metric. This one went from 5.4% to 14.1% in the three weeks after the title was rewritten, and the top reason changed with it — from “no longer needed” to “not as described”. Nothing about the product changed. Seventy-two per cent of them come back sellable, which is the signature of an expectation problem rather than a fault: people are opening the box, finding something other than what they thought they bought, and sending it back unused.
|
| ● |
PB224 |
Therapy putty, 4-pack |
3,120 |
74 |
2.4% |
3.8% |
falling |
no longer needed |
$888 |
$1,704 |
$23.00 |
12% |
cannot be resold once opened |
| ● |
SW043 |
Baby scale weighing tray |
1,240 |
108 |
8.7% |
6.9% |
rising |
defective or broken |
$3,240 |
$6,120 |
$56.70 |
9% |
a real fault — same crack, 3 photos |
|
The refund is roughly half of what a return costs. The outbound fulfilment fee is spent and does not come back. The return has its own fee. Somebody inspects it, and most units cannot go out again as new — putty is 12% resellable, a scale tray with a crack is 9%. So the True cost column is refund plus both fees plus the written-down unit, and it is about double the refund line every time. A returns report built on refunds alone understates the problem by half, which is why nobody acts on it.
|
| ● |
PB211 |
Sensory bin scoops, 6-pack |
980 |
18 |
1.8% |
4.1% |
steady |
no longer needed |
$204 |
$392 |
$21.80 |
61% |
nothing |
| ● |
PB038 |
Coil spring walking toy |
1,910 |
41 |
2.1% |
4.1% |
steady |
arrived damaged |
$486 |
$948 |
$23.10 |
22% |
packaging — 4 from one FC |
|
And the rate on any young month reads low, for the same reason profit does. A return arrives about three weeks after the sale, so a month that is half over shows returns against sales that have not had time to come back. The rate is calculated against the cohort that has aged, not against the calendar month, and the header says which window it used — otherwise every month looks like an improvement until it is closed.
|