Published
Shipping savings get quoted two ways, and the two numbers feel completely different even though they are the same number. Per shipment it sounds like rounding. Per month it sounds like a headline. Both are just volume multiplied by the per-order difference.
So here is the per-order figure, what it does and doesn’t cover, and why the label price is usually the smaller half of the answer.
This section is about one thing only: buying a cheaper label for the same parcel and the same delivery date, by choosing the right box and the right service per order. Nothing else. It is the easiest part to quantify, which is why it gets quoted, and it is not the part that decides whether an engagement is worth doing.
Across the operations we work with, that piece falls into two bands, and volume is the rough dividing line.
Under about 6,000 orders a month: $1 to $3 per order. Order profiles at this volume tend to be more uniform. Fewer SKUs, fewer package sizes, fewer carrier accounts connected. The cheapest service is often the obvious one, and a well-configured set of automation rules is already capturing part of what is available.
Above about 6,000 orders a month: $3 to $5 per order. Not because the software works harder, but because the operation has more room in it. More SKUs in more sizes means more orders where the box is genuinely arguable. More carrier relationships means more services actually competing. Heavier and more varied parcels means the spread between the right answer and the default answer is wider on each one.
The arithmetic from there is unremarkable, and multiplication is not the interesting part. What matters more is that this band is a floor rather than a total.
The intuition people usually bring is that a bigger operation is a more optimised one, so there should be less left on the table. In practice the opposite tends to be true, for a specific reason.
Savings come from decisions where the right answer varies order by order. A catalogue of six SKUs in two box sizes does not have many of those — the right box is nearly always the same box. A catalogue of six hundred SKUs across nine box sizes has one on almost every multi-item order, because the combination of items in the order is what decides the carton, and that combination is different every time.
The same applies to rating. One carrier connected means very little to compare. Five carriers, some with negotiated rates and some without, means a real comparison on every parcel, and the comparison gets more valuable as parcels get heavier, because that is where service tiers diverge most.
So complexity, not size, is what produces the saving. Volume just correlates with complexity, which is why it works as a rough dividing line.
For an operation with real complexity, the label price is the part that is easiest to measure and rarely the part that matters most. The rest of it does not appear on a carrier invoice, which is why it goes unquoted and why it is usually larger.
Manual labour. The decisions a person makes order by order because no rule expresses them — which carton, which service, which orders go in which batch. That time is a standing cost, it scales with volume, and it is paid whether or not the decision was right.
Exception handling. The orders that fall out of the normal flow and need someone to look at them. Each one is small. The queue is not, and it tends to land on whoever is most experienced, which is the most expensive way to resolve it.
Spoilage on temperature-sensitive shipments. Where the packing decision depends on the forecast at the destination and the transit time, getting it wrong costs the product, the reship and the customer, not a few cents of postage.
Split shipments and re-picks. An order that goes out in more boxes than it needed, or one that comes back to the floor because the packer could not make what was picked fit what was chosen. Both are labour, both are packaging, and the second is a second label.
Consistency that does not depend on who is working. The same order gets the same decision on a Tuesday in March as it does the week before Christmas with temporary staff on the floor. This is the hardest to put a number against and often the one operations say they felt first.
None of these are quoted per shipment because they are not per-shipment costs. But an operation shipping 10,000 orders a month with a person making packing calls all day is not primarily buying cheaper labels.
These are ranges from operations we have worked with, and they are ranges for a reason. Two businesses at identical volume can be several dollars an order apart depending on how varied their catalogue is, how many carrier accounts they have, how heavy their parcels are, and how much of the available saving their existing automation is already capturing.
Which is why we do not ask anyone to take the range on faith. Before an engagement starts, we run your actual order history through the logic we would build and show you what it would have done differently on your own orders — which box, which service, what the difference was. You see the number for your operation before you are paying for anything.
If that simulation comes back small, that is the answer, and we say so. An operation whose existing rules already express its logic well will not have much of a gap, and there is nothing we can build that changes that.
The label half is not a windfall. It is a permanently lower cost per order, which is a different and more useful thing: it compounds with volume, it improves the arithmetic on paid acquisition, and it makes free shipping thresholds and margin on lower-priced SKUs viable at levels that did not work before.
The other half shows up as capacity. The same team handles more volume without the decisions getting worse, and peak stops being the period where everything that was held together by one person’s attention comes apart.
If you want the number for your own operation, the self assessment works it out from two standard ShipStation exports. It runs entirely in your browser, your order history never leaves your machine, and there is no call attached to it.
Knowing how a carrier prices a parcel is useful. The harder question is which parcel to make — because the packing decision sets the dimensional weight and the cubic tier, and the cheapest box changes with the contents, the destination and the rate card.
There are eight situations where that decision has grown past what a rule can express. Most operations are in two or three of them.
