Importing Parts: Allocating a Container Honestly
A mixed container of heavy-and-cheap alongside light-and-expensive is the exact case where freight allocation matters most and where the correct basis is not available. What to do instead, the split-delivery defect to work around, and why demurrage always lands after you have started selling.
Importing parts into Kenya produces a specific and expensive confusion: you know exactly what the container cost and you do not know what anything in it cost. One freight invoice, one duty assessment, one clearing bill, and two hundred item lines that each need a share of them. Get that spread wrong and your margin figures are wrong per line while being right in aggregate — which is the worst kind of wrong, because it looks fine on the profit and loss and quietly misdirects every pricing and reordering decision you make.
The allocation problem, stated plainly
Two bases are available: by value and by quantity. A parts container is usually the one shipment shape where both are wrong.
The same container, allocated two ways
Quantity allocation is right for a shipment of similar items — a container of nothing but filters — and catastrophic for a mixed one, because a five-shilling clip cannot absorb a share of freight equal to a radiator's. Weight or volume would be the correct basis for a mixed container and is not available, so allocate by value and then manually correct the handful of lines you know are distorted: anything heavy and cheap is under-costed, anything light and expensive is over-costed.
The split-delivery problem
This one is a genuine defect rather than a scope limit, and it is worth knowing because parts imports split constantly — part of a container clears, the rest follows a week later.
Landed cost allocation finds the batches for a purchase order through the most recent check-in against that order. So if a shipment arrives in two deliveries and you allocate the cost after the first, the whole freight and duty burden can land on that first delivery's batches — leaving the second delivery costed at invoice price alone and the first over-costed by roughly double.
How to work around it
- Allocate after the final delivery, never after the first. One rule, and it removes the problem entirely for a shipment you know is splitting.
- If you must allocate early, raise the deliveries as separate purchase orders with their own share of the costs. More admin and it keeps each allocation attached to the goods it belongs to.
- Read the resulting unit costs before pricing anything. A unit cost that has roughly doubled against your expectation is the signature of this problem, and it takes thirty seconds to spot at the moment it is cheap to fix.
- Note which orders you split. In six months you will not remember, and the unit costs will look like a mystery rather than a known cause.
[Demurrage](/glossary/demurrage), and why the timing hurts
Duty and freight are known reasonably early. Demurrage, storage and any final clearing adjustment are not — they arrive after the goods have left the port, which in practice means after you have started selling from the shipment.
And there is no re-costing of stock already sold. So the late cost lands entirely on what remains, which produces two distortions in sequence: the early sales from a container look more profitable than they were, and the remaining stock looks more expensive than it is. Both are artefacts of timing rather than of trade.
Best
Hold the shipment for a week before selling
Commercially painful and it lets the full cost land before the first sale. Worth it on a high-value shipment where the margin decision matters; unrealistic on fast-moving lines your customers are waiting for.
Practical
Add a demurrage provision to your allocation up front
A standing percentage based on what your last six shipments actually incurred. Wrong on each individual shipment and much closer than zero, which is what a nil provision assumes.
Minimum
Treat pre-allocation margins as provisional
Mark it clearly in your own head: any margin figure on a shipment whose costs are not final is an estimate. Do not set a long-term price from it.
Avoid
Allocating the late cost across remaining stock and forgetting
This is the default and it produces a subset of your item list carrying a cost that reflects a delay rather than a value. Six months later somebody wonders why that line never sells.
The currency layer
Three exchange rates matter on any import and they are all different: the rate on the order date, the rate on the payment date, and the rate when the goods cleared. There is no FX handling inside landed cost — no payment-date rate, no weighted rate across staged payments, no per-shipment variance — so this discipline is entirely yours.
The practical rule is short: record what you actually paid in shillings on the day you paid it, and treat any conversion at invoice date as an estimate rather than a cost. Where a shipment is paid in stages, the true rate is a weighted average you compute — and if the shilling moved between the deposit and the balance, the difference is real money that belongs in the unit cost. Paying in shillings versus dollars covers the same mechanism from the software-buying side.
What is built for imports
What AWRA OpsHub does today
- Landed cost types on a purchase order, allocated across the shipment by value or by quantity.
- Allocation reaching the batch, so a specific batch of a part carries its own landed cost rather than an average.
- Landed cost rolling into weighted average cost, which every margin figure, journal posting and valuation reads.
- Two-way matching at receipt — ordered against received — with an over-receipt guard before the adjustment is written.
- Attachments on the purchase order, so the bill of lading, IDF, entry and clearing invoice live on the record.
- Serial and batch tracking, so a specific physical unit traces back to the shipment and cost it arrived with.
What it does not do
- No weight or volume allocation basis — the one that would be correct for a mixed container.
- Split deliveries allocate against the most recent check-in only, so allocating before the final delivery mis-assigns the whole cost. This is a defect, not a design choice.
- No FX handling inside landed cost of any kind.
- No re-costing of stock already sold when a cost arrives late.
- No customs or clearing-agent data import — every figure is entered by hand from the documents.
- No duty calculation. Nothing computes what you should owe; you record what you were assessed.
The one habit that matters most
Do the landed cost calculation on every shipment, even roughly, even late. A business that allocates approximately on all twelve containers a year knows its costs far better than one that allocates perfectly on two and prices the other ten off invoice value. Approximation applied consistently beats precision applied occasionally, and this is the clearest example of it in the whole trade.
Our take
Allocate by value, then hand-correct the heavy-and-cheap lines, because the basis that would be right for a mixed container is not available. Never allocate before the final delivery of a split shipment — that is a real defect and one rule avoids it. Carry a demurrage provision based on your own last six shipments rather than assuming zero. And record what you actually paid in shillings on the day you paid it, because nothing in landed cost handles exchange rates for you.
Know what the container actually cost
Landed cost allocated to the batch and rolled into one [weighted average cost](/glossary/weighted-average-cost) that every report reads — with the allocation bases, the split-delivery caveat and the absence of FX handling stated plainly.
See plans & pricingFrequently asked questions
How should freight and duty be allocated across an imported parts shipment?
By value, in almost every case. Quantity allocation is correct for a container of similar items and catastrophic for a mixed one — 5,000 clips would absorb 99% of the freight and a five-shilling clip cannot carry a radiator's share. Weight or volume would be right for a mixed container and is not available, so allocate by value and hand-correct the heavy-and-cheap lines, which will be under-costed.
What happens if a shipment arrives in two deliveries?
Allocation resolves batches through the most recent check-in against the purchase order, so allocating after the first delivery can put the entire freight and duty burden on that delivery — over-costing it by roughly double and leaving the second delivery at invoice price alone. This is a defect rather than a scope limit. Allocate only after the final delivery, or raise the deliveries as separate purchase orders.
How do we handle demurrage that arrives after we have started selling?
There is no re-costing of stock already sold, so a late cost lands entirely on what remains — early sales look more profitable than they were and remaining stock looks more expensive than it is. Carry a standing demurrage provision based on what your last six shipments actually incurred; wrong on each shipment individually and much closer than assuming zero.
Does landed cost handle exchange rates?
No. There is no payment-date rate, no weighted rate across staged payments and no per-shipment variance. Record what you actually paid in shillings on the day you paid it, and treat any invoice-date conversion as an estimate. Where a shipment is paid in stages and the shilling moved between deposit and balance, that difference is real money that belongs in the unit cost.
Does the system calculate import duty?
No. Nothing computes what you should owe — you record what you were assessed, from the entry. There is also no customs or clearing-agent data import, so every figure comes off the documents by hand. What the system does is spread those figures across the shipment and carry them into unit cost.
Is it better to allocate roughly or not at all?
Roughly, always, on every shipment. A business that allocates approximately on all twelve containers a year knows its costs far better than one that allocates perfectly on two and prices the other ten off invoice value. Consistency beats precision here more clearly than almost anywhere else in the trade.
