Teams that stopped arguing with their stock file

Nordwind Outfitters sells technical outerwear and camping equipment through six distribution centres and retail doors. Its demand is brutally seasonal: roughly of annual revenue lands in fourteen weeks. Before Stockmere, each DC ran its own stock file and reconciled to the ERP in a nightly batch. By the time a September morning report said a parka was in stock in the eastern DC, it had often been promised twice, once to a store replenishment order and once to an e-commerce customer.

The operations team implemented Stockmere in eleven weeks, starting with a two-DC pilot. Every receipt, pick, transfer, and adjustment now posts to a single event ledger within seconds, and available-to-promise is computed against that ledger rather than against yesterday's snapshot. Rolling cycle counts replaced the annual wall-to-wall count: Stockmere schedules daily count tasks weighted by velocity and variance history, so the highest-risk locations are counted most often.

The first peak season on the platform told the story. Stockout events, meaning an order line that could not be fulfilled from the promised location, fell year over year. Record accuracy, measured by blind count audits, rose from to . With trustworthy records, planners cut safety stock on the top two velocity bands, taking out of carrying cost without hurting fill rate.

“We used to argue about whose spreadsheet was right. Now there is one number, it's current, and the count program keeps it honest. Peak season planning starts from facts.”

Ferrous Industrial distributes fasteners, bearings, and certified steel stock to manufacturers across eleven sites. Much of its catalogue is lot-tracked: when a customer machining aerospace brackets asks where a batch of bar stock came from, "somewhere in receiving, probably March" is not an acceptable answer. Yet that was roughly the fidelity its legacy system offered. Adjustments were untyped, transfers lost their history, and the annual audit meant three weeks of assembling paper trails.

Stockmere's append-only movement ledger changed the shape of the problem. Every receipt is tied to a supplier lot; every pick, transfer, and adjustment references the lot and the user who made it, with a mandatory reason code on write-offs and corrections. The compliance team built saved trace queries ("show every movement of lot X, with documents") that export in minutes.

Fourteen weeks after kickoff, all eleven sites were live. The next external audit was prepared in four days instead of three weeks, because the evidence was a query rather than a scavenger hunt. Reason-coded adjustments exposed where product was leaking, a mis-labelled scrap process at two branches, and shrink fell from to of cost of goods within three quarters.

“Traceability used to be a project we ran every autumn. Now it's a property of the system. The auditors asked for our sampling method, ran their queries, and left early.”