The stock was not missing. It was in the wrong place.
A catalogue too big to hold in anyone's head. Over eight hundred live ASINs moving from Chinese factories into more than a hundred Amazon fulfilment centres, with nothing joining the two ends together. We built the file that joins them, then ran the weekly cycle off it.
Overstocked at one node, short at the other.
As one number the catalog looked oversupplied. Read node by node it was oversupplied and short at the same time.
A 177-day lane, named step by step.
This is not a supplier quote. It is 26 steps with days on each, set per supplier and per freight mode.
Where the stock sits, and what else the work covers.
Stock is counted in five separate places. A single total would hide every one of them.
Inbound shipments from China into 119 distinct Amazon fulfilment centres plus AWD receiving, across 18 months, each carrying status, ETD, ETA and a tracking number on all but one line.
Factories mapped to the products they make, with days-of-stock policy set per node rather than as one blanket number: 45 to 90 days at FBA, 30 to 120 at AWD, 60 to 220 in total.
SKUs reviewed weekly, with a written instruction issued on 846 of them across both FBA and AWD. Not a flag. An instruction that reads “send 150 days of stock to FBA” with the quantity already worked out.
FBA, AWD, the New Jersey warehouse, ready-to-ship stock sitting in China, and units still in production. 79,871 units in New Jersey, 41,089 held ready in China across 103 SKUs, 11,056 still on the line.
On-hand inventory value across 118,363 units, tracked at both FBA and AWD for every SKU.
A liquidation manifest covering 336.6 pallets, 92,315 units and $1.93M of retail value, with a photograph or a live listing attached to most lines.
If your total looks healthy but the selling node keeps running low, start with the map.
Draw your own nodes and lanes in the free map tool, or bring a stock-by-location report to a 30-minute call.