Reference
Consigned stock, and
what your system should say
Consignment is stock that sits in one party's building and belongs to another. The goods move first and the ownership moves later — usually at the moment of sale — and that gap is the whole of it.
It is an ordinary commercial arrangement and a genuinely awkward one to hold in software, because almost every stock system is built on the assumption that what is on your shelf is yours. Consignment breaks that assumption in one direction or the other, and most systems have no field for the difference.
It runs in two directions, and the word does not distinguish them
Ask two merchants about consigned stock and they will describe opposite arrangements with the same word. Establishing which one is in play is the first question, because the accounting, the risk and what the stock system has to say are different in each.
You hold stock somebody else owns
A supplier's goods in your warehouse or on your shop floor. You can sell it, you did not buy it, and you owe for it once it goes. It is not your asset and it should not sit in your stock valuation — but it absolutely must sit in your availability, because you can sell it today.
This is the direction that flatters a balance sheet by accident when a system cannot tell the difference: stock value goes up, nothing was bought, and margin reads wrong until somebody reconciles it by hand.
Somebody else holds stock you own
Your goods in a retailer's shop, a distributor's warehouse, a marketplace's fulfilment centre, or with a 3PL. It is your asset, it is your risk, and it is nowhere you can count it.
The exposure here is the opposite one: stock you own is invisible to your own availability, so you either understate what you can sell or you double-sell it — depending entirely on whether anything is feeding you the other party's numbers, and how often.
Four moments, and only one of them is the sale
A consigned unit passes through these in order. Every argument about consignment — who insures it, whose stocktake it appears on, when it becomes revenue, who wears it when it is damaged — resolves to which of the four the contract says title moves at.
-
1 — Dispatch
Goods leave the owner. Nothing has been sold and nothing has been bought. If a system books this as a sale, every figure downstream is wrong from here — and it is a common default, because dispatch is the event the warehouse naturally raises.
-
2 — Receipt into the holder's building
Now it is physically present somewhere it is not owned. This is the state most stock systems cannot represent, and the workaround — a separate location, a flag, a second SKU — is a decision worth making deliberately rather than discovering.
-
3 — Sale to the end customer
Usually where title passes, and usually twice in the same instant: owner to holder, holder to customer. Two transactions, one event, and if only one of them is recorded the two parties' books stop agreeing from that day.
-
4 — Settlement, or return
The holder pays for what sold, and unsold goods go back or roll into the next period. The reconciliation here is against a statement the other party produces, which is why consignment quietly becomes a monthly manual job in almost every operation that has not designed for it.
Where title passes and how VAT follows it are contract and tax questions, not software ones — they belong with the agreement and whoever advises on it, and this page does not answer them. What a system can do is represent whatever the contract says faithfully, which it cannot do until somebody has read the contract and said which moment is moment three.
Where it actually goes wrong
None of these are consignment being difficult. They are all the same root — a system holding one number for stock when consignment needs two, ownership and availability, and they are not the same set.
The valuation includes what you do not own
Held consignment stock counted as an asset. The stock figure the ledger takes at period end is overstated by whatever is on the floor unsold, and nothing in the software objects, because to the software it is simply stock.
Your own stock is invisible to your own channels
Stock you own sitting in a third party's building, with no feed back. You cannot sell it because you cannot see it, or you sell it twice because your only view is a spreadsheet somebody emails weekly.
The stocktake finds goods nobody can classify
A count that cannot separate owned from held produces a variance that is not a variance, and a fortnight of investigation into a discrepancy that was never real. It repeats every count until the flag exists.
Two parties' books stop agreeing
The settlement statement and your own sales record diverge, usually over returns and timing, and the reconciliation becomes a monthly negotiation. It is the same shape as the payouts-against-orders problem on the accounting reference, with a trading partner in place of a payment processor.
The system requirement that falls out of all four is a single one: ownership and availability have to be separately expressible per unit, per location. Which system is nominated to hold that, and the four numbers any candidate needs to show, is stock control software; which system decides what a consigned unit is promised to is the order layer.
What this reference does not yet contain
We have not delivered a consignment implementation on Shopify Plus, and this page claims none. What is above is the structure of the problem and the questions that decide an architecture — which is the part that generalises, and the part a vendor cannot write.
What is not here: how each stock product actually represents held-not-owned goods, which of them can carry a second ownership dimension without a second SKU, and what the settlement reconciliation costs to run per month. Those are implementation findings and they appear named and dated when there are some.
If consignment is already in the operation, the useful first read is narrow: which moment the contract says title passes at, and whether anything in the stack can currently express ownership separately from availability.
Book the diagnosticTell us what is breaking
What the systems are doing now, and what you need them to do. We will tell you whether it is a platform problem or something cheaper.