The invoice that arrives is not the cost that matters
Rent is the number everyone knows because it turns up monthly with a building attached to it. Run the defaults here and rent is roughly a third of what the warehouse costs to operate. Labor is larger. Add equipment and the amortised racking and the storage side — the part that would still cost money if nothing moved all month — is somewhere under half.
That ratio is the reason a single cost per position is a useful summary and a poor pricing instrument. Two customers occupying a hundred positions each can consume wildly different amounts of the total, because one turns their stock weekly and the other stores a seasonal item that arrives once and leaves once. A blended rate charges both the same and loses money on one of them.
Occupied positions, not positions
The cost is fixed against the positions you built, and the revenue or the value comes from the positions holding something. At 88 percent occupancy the cost per occupied position is about 14 percent higher than the headline figure, and the gap grows fast as occupancy falls — at 70 percent it is over 40 percent higher.
This is where the temptation to run at very high occupancy comes from, and it is worth resisting past a point. A warehouse at 95 percent occupancy has no room to receive a large inbound, no room to consolidate, and no room to hold anything while it is counted, so the labor cost of every move rises to pay for the storage saving. The right occupancy target is the one where the two curves cross for your operation, and it is usually lower than the number that looks efficient on a spreadsheet.
Which lines actually move
| Line | How fixed it is | What changes it |
|---|---|---|
| Rent | Fixed for the lease term | Renewal, or using the height you already lease |
| Racking | Sunk, spread over years | Only at the point of reconfiguration |
| Utilities and building | Semi fixed | Lighting, heating schedules, dock seals |
| Labor | Variable with volume, sticky downwards | Slotting, batching, layout, volume itself |
| Equipment | Steps rather than slopes | Fleet size at the point of renewal |
Read down the right-hand column and it becomes obvious why the same two projects come up in every warehouse. Using the height you already rent attacks the largest fixed line without changing the lease, which the warehouse cube utilization calculator quantifies. Reducing travel attacks the largest variable line, which the order picking travel time calculator puts in hours and the slotting calculator puts into a move list.
Using this to price work for other people
If the output is going into a quote, split it before it goes anywhere near a customer. Storage cost per position per month covers rent, building and racking, and it is genuinely a per-position-per-month thing. Handling cost per pallet moved covers labor and equipment, and it is genuinely a per-move thing. Quoting them separately is not an accounting nicety, it is the only structure that survives a customer whose volume changes.
Then add whatever margin the work is worth, remembering that this calculator produces cost and nothing else. It has no view on what anyone should charge, no market rate built in, and no idea what your competitors do. The break even calculator handles the volume side of that question and the employee total cost calculator is worth running before you trust the wage figure you put into the labor field, since the fully loaded number is reliably higher than the one people have in mind.
Questions people ask
What is a normal cost per pallet position?
No figure worth quoting exists, and this page deliberately ships without one. The result depends on rent in your market, the age and height of the building, how many positions the layout achieved per square foot, wage levels, how much handling the stock demands, and whether racking was inherited or bought last year. Two operations a mile apart can differ by a factor of two on identical freight. The number that means something is your own, calculated from your own inputs and tracked against itself over time.
Should empty positions be costed?
They already are, which is the point of showing the occupied figure separately. Rent, racking and most of the building cost are incurred whether a position holds anything or not, so the total divides across every position you built while only the occupied ones do any work. Some operations then allocate the empty-position cost across the occupied ones, which produces the higher per-occupied figure here and is the honest number for pricing. Others treat it as the cost of available capacity and hold it centrally. Both are defensible; what is not defensible is quoting the lower number and being surprised at the year end.
Why is handling separated from storage?
Because they scale with completely different things and mixing them hides that. Storage cost follows how long a pallet stays; handling cost follows how many times it is touched. A blended rate per position charges a fast turning customer and a slow one the same, which means one of them is subsidising the other and nothing in the reporting reveals which. Third party warehouses price the two lines separately for exactly this reason, and the split is just as useful internally, since it tells you whether a cost problem is a building problem or a labor problem.
Is amortising the racking the same as depreciating it?
No. Spreading an installed cost over a chosen number of years is an estimating device for comparing operating scenarios month to month. Depreciation is an accounting treatment with rules about capitalisation, useful life and method, and racking installed in a leased building is often treated as a leasehold improvement tied to the lease term rather than to the physical life of the steel. Those two treatments produce different monthly numbers, and the accounting one is the one that belongs in your books. Use this for decisions, not for reporting.
How do I count pallets handled?
Count every touch that consumes labor, not every pallet in the building. A pallet received and later shipped is two moves. If it is put away, replenished to a pick face and then picked from, it is more than two, and if your operation does a lot of that, counting only receipts and shipments will understate the handling cost per move substantially. The cleanest source is your system's transaction count rather than an estimate, because the moves people forget are exactly the internal ones that make up the difference.