The Invoice That Arrives After the Goods Are Gone
Your clearing agent bills you three weeks after the container cleared. By then some of the goods have sold. That freight and duty will attach to whatever is still on the shelf and to nothing that has already gone — so the margin you reported on the fastest-moving lines is the one that was most wrong.
Landed costing has an assumption inside it that nobody states, because in a textbook it is always true: that you know what the goods cost to land before you sell any of them.
The verdict
This is the kind of gap that never announces itself, because nothing fails and the totals reconcile. Costs do all land eventually, so the period looks right; the damage is entirely in the comparison between products, where your fastest lines flatter themselves and your slowest carry a charge that was never theirs. If your suppliers invoice with the goods, none of this reaches you. If you import through a chain where the clearing account arrives weeks later and the stock sells in days, start estimating charges at receipt — and ask us, and everyone else, what happens to the sold half when the invoice finally turns up.
Importers know it is frequently false. The goods clear and go to the shelf. The freight invoice, the clearing agent's account, the port charges and the duty receipt arrive over the following weeks, in no fixed order, from several parties, some of whom bill monthly. Meanwhile the fastest lines are already selling — which is precisely why you imported them.
What our system does when the cost arrives late
It attaches the cost to the goods and recalculates the average cost of what you are holding. That figure then applies to everything sold from that moment on.
It does not reach backwards. A sale that has already been posted keeps the cost it was given at the time, which was a cost that did not yet include the freight. Nothing restates it, and nothing flags it as having been costed on incomplete information.
-
The container clears and the goods are received
Unit cost is the purchase price plus whatever charges you already know about — often only the freight quote, sometimes nothing at all.
-
The fast lines start selling
Each sale is costed at the figure standing at that moment, and a margin is reported against it. The margin looks good, because a real cost is missing from it.
-
The clearing invoice arrives three weeks later
It is attached to the order and allocated across the goods, and the average cost of the remaining stock rises to absorb the whole charge.
-
The correction lands entirely on what did not sell
The slow lines now carry the freight for the fast ones. Nothing that has already gone is touched, and no report says so.
The cost correction lands on whatever is left, which means your slowest stock absorbs the import cost of your fastest.
That last step is the part worth sitting with. It is not simply that the early sales were under-costed — it is that the error is then concentrated onto exactly the lines that did not sell, making them look less profitable than they are, while the lines that did sell look better than they were. Both halves of your product ranking are distorted, in opposite directions, by an accounting timing difference rather than by anything commercial.
Why this is sharper across the islands
Two things have to be true for this to bite: import charges have to be a meaningful share of unit cost, and they have to arrive late. Across this region both are routine — goods reach you through a transshipment chain with several parties billing separately, and the agent's account commonly follows the clearance by weeks rather than days.
Add the third factor, which is that a small market sells through a consignment quickly, and the window between the goods arriving and the cost arriving is frequently longer than the window between the goods arriving and the goods leaving. The costs are chasing stock that has already gone.
What the timing difference does to figures you rely on
What you read What was true
Margin on the fast-selling lines Overstated
They were costed before the freight and duty landed, and nothing went back to correct them.
Margin on what is left Understated
The remaining stock absorbed the entire charge, including the share belonging to units already sold.
Which products are worth importing Ranked wrongly
The distortion runs in opposite directions on the two groups, so the gap between them is exaggerated twice over.
Total gross profit for the period Approximately right
The costs do all land eventually. This is the reassuring part, and it is why the problem survives — the total reconciles while the per-line figures do not.
The bottom line being roughly correct is exactly what stops anybody investigating. Nothing fails, nothing is out of balance, and the distortion lives entirely in the comparison between products rather than in any total.
Why we are not simply going to fix it
It would be easy to write this as a defect and promise retrospective re-costing. We would rather explain why that is a harder call than it sounds.
Restating the cost of a transaction that has already been posted means changing a number that other things have been built on — a period that may be closed, figures somebody has already reported, and in some cases a ledger entry. A system that quietly rewrites history is worse than one that does not, because you lose the ability to say what any report meant on the day it was produced. Our position, stated in our costing copy, is that a correction is a new transaction rather than an edit to an old one, and that is a defensible position rather than an oversight.
What is not defensible is doing that silently, which is what happens today. There is no marker on stock costed provisionally, no report of orders where costs are still expected, and no way to see how much of a period's margin was calculated on incomplete information. The design decision is sound; the absence of any visibility around it is the actual gap.
What to do about it now
Estimate the import charges at receipt rather than waiting for the invoice. Attaching an approximate freight and duty figure at receipt gets most of the cost onto the goods before they sell, and the later invoice becomes a small correction rather than the whole charge landing on the remainder. It is less accurate in the ideal case and far more accurate in the case that actually happens to you.
Two, and neither of them rewrites history
We would keep the forward-only rule and make it visible, which we think is the right trade rather than the easy one.
A provisional cost marker and an expected-costs list
Mark stock received against an order that is still awaiting charges, and list those orders with what is expected. This changes no arithmetic and tells you which figures are incomplete, which is the information you actually lack.
Estimate at receipt, reconcile at invoice
Record an expected freight and duty figure at receipt so goods are costed close to correctly before they sell, then compare the estimate against the invoice when it lands. Most of the distortion disappears, and the comparison tells you how much to trust your own estimates.
The second is the one that fixes the problem in practice. The first is the one that tells you the problem is happening, which is why we would ship it first.
Talk to us about import costingWhat AWRA OpsHub does today
- Costs attached to an order at any point and allocated across the goods, whenever the invoice arrives.
- Allocation that totals the bill across any number of partial deliveries, in proportion to what each holds.
- The allocation retained per consignment, so what one arrival cost to land is recoverable afterwards.
- The weighted average updated from it, so everything sold after the correction is costed properly.
What it does not do
- No re-costing of goods already sold. The correction moves forward only; posted sales keep the cost they were given.
- No provisional cost marker, so stock costed before its charges landed looks identical to stock fully costed.
- No expected-costs view, so orders still awaiting a clearing invoice cannot be listed.
- No estimate-versus-actual comparison, so how good your estimates are is not measurable.
- No flag on margin reports that a period included provisionally costed sales.
Not ours, by choice
- We publish no port charge, tariff or clearing fee for any market in this region. That agents commonly bill weeks after clearance is a description of commercial practice, not a rate we are sourcing.
- We are not describing forward-only correction as a bug. It is a deliberate choice with real reasons behind it, and the gap we accept is that it happens without any visibility.
Four questions about costing goods that sell before their invoices arrive
Sell half a consignment, then add the freight invoice. What happens to the sold half?
What a straight answer sounds like
A restatement, or nothing. Ours is nothing.
Why it matters
This is the whole question and it takes one demonstration. The answer determines whether your margin reports mean anything per line.
Can I see which stock was costed before all its charges arrived?
What a straight answer sounds like
A marker or a report. Ours has neither.
Why it matters
Provisional and final costs look identical, so nobody knows which figures to distrust.
Which orders are still waiting on a clearing invoice?
What a straight answer sounds like
A list. Ours cannot produce one.
Why it matters
These are known unrecorded costs. Not seeing them means every margin report is optimistic by an unknown amount.
Can I estimate charges at receipt and reconcile later?
What a straight answer sounds like
Yes, plus a comparison of estimate against actual.
Why it matters
Estimating is the practical answer to this whole problem, and measuring the estimates is what makes it improve.
Send us a consignment and its charges as they actually arrived
With the dates. We will show you what the goods were costed at while they were selling, what they should have been, and how much of the difference an estimate at receipt would have removed.
Talk to us about landed cost