Profit Per Purchase Order: Which Batches Made Money
Track each unit at the cost of the purchase order it came in on, landed costs included, and see which batches made money and which sat on the shelf.
To see which purchase orders made you money, give every unit the cost of the purchase order it came in on, including its share of freight and other landed costs, and track it by serial number. When the unit sells, the serial tells you exactly which purchase order it came from, and that sale's profit goes back to the right order. A purchase order's profit is then the total over the units sold so far, and its sell-through tells you how much is still sitting on the shelf. An average cost across all your buys can't answer the question, because it blends your good buys with your bad ones.
Why does an average cost hide a bad buy?
Most tools keep one cost per product: the average of everything you've paid. That number is fine for valuing a warehouse. It's useless for deciding whether to place the same order again. If one supplier's batch came in cheap and the next one came in expensive, the average says both are making the same money. One of them isn't.
What counts as the cost of a unit?
The price on the supplier's invoice is only part of it. Anything you paid to get that batch onto your shelf belongs in the unit cost:
- Inbound freight, including expedited shipping you paid to hit a deadline
- Duty, customs brokerage and port fees on imports
- Anything you paid to prep, test or repair the units before they could sell
Spread those costs across the units on that purchase order. Split by quantity when the units are similar; split by value when one cheap item and one expensive item share a container, or the cheap item ends up carrying freight it didn't cause.
What does that look like on two real-sized orders?
Two purchase orders for the same product, both sold on Amazon at $30. The figures are an invented example; the shape is common.
| Line | Purchase order A | Purchase order B |
|---|---|---|
| Units bought | 100 at $10.00 | 100 at $12.00 |
| Freight | $150 (standard) | $400 (expedited) |
| Real unit cost | $11.50 | $16.00 |
| Marketplace fee + label per sale | $9.50 | $9.50 |
| Profit per unit sold | $9.00 | $4.50 |
| Units sold so far | 90 | 40 |
| Profit so far | $810 | $180 |
| Sell-through | 90% | 40% |
On a single average cost of $13.75, both orders would show $6.75 a unit and look the same. Split by purchase order, the second one earns half as much per unit and is selling slower. The rush freight was the expensive decision, and you only see that once the cost follows the unit.
What happens with returns and write-offs?
A return that goes back on the shelf goes back into that purchase order's stock at its original cost, and the sale's profit comes off. A unit you can't resell is a loss at what you paid, and it counts against the purchase order it came from. That's how a supplier with a high defect rate shows up as a bad supplier instead of a slightly worse average.
How do I do it by hand?
- Give every purchase order its own tab: units, invoice cost, freight and landed costs.
- Divide the landed costs across the units to get a real unit cost.
- Tag each sale with the purchase order the unit came from. Without serial numbers you're guessing here; with them, the serial on the order settles it.
- For each sale, subtract the marketplace fee, the label and the unit cost. Sum by purchase order.
- Track units sold against units bought for sell-through, and count write-offs against the order.
Why do serial numbers make it exact?
Once two batches of the same product are on the shelf, every sale has to come out of one of them. Without serial numbers, nobody knows which, so the profit gets split by a rule: oldest first, or an average. That's an estimate. With a serial on every unit, the order records the exact unit that shipped, the unit knows its purchase order and its cost, and a purchase order's profit is the sum of real sales, not a guess. If you sell anything where batches differ in cost or quality, serials are what turn this from an estimate into a number you can act on.
Do I have to scan serials on every move?
No. Full serial tracking means scanning the unit at every step: receiving, every bin move, every warehouse transfer, picking and shipping. That's worth it for high-value stock, and it's a lot of scanning for everything else. For profit per purchase order you only need two scans: when the unit is received against the purchase order, and when it ships on an order. Those two ends tie every sale to its batch. What happens on the shelf in between doesn't change the cost.
For what the rest of a sale costs, see true profit per order on Amazon, Walmart and eBay.
How Rilk shows it
Freight, duty, brokerage and other landed costs you add to a purchase order raise the unit cost Rilk uses for profit, not just the invoice price. Track a product by serial number and each unit sold is costed from its own serial, so every sale is tied to the exact purchase order it came from. You choose the depth per product: require a serial only at receiving and at shipping (the light version, no scanning on bin moves or transfers), or turn on full tracking where you want every move recorded. Without serials, units of the same product are costed at their average, which is fine for interchangeable stock and an estimate for anything else.
Every purchase order then has its own profit report: units bought and sold, profit, repair cost and sell-through. The purchase-order profit report and true per-order profit are on Pro and Ultimate, and every account gets one 14-day Pro trial to run them on your own orders.
Start free at rilk.ai — no credit card required. See plans.
About the author
Nouri Shaaya
CEO, Rilk
Nouri Shaaya is the CEO of Rilk. He has been selling online since 2013, running a multi-channel operation across Amazon, Walmart, eBay and retail drop-ship, with sales approaching $40 million a year.
More from NouriReady to simplify your operations?
See how Rilk replaces your spreadsheets and duct-taped tool stack with one platform built by sellers, for sellers.
