Profit Radar scores every order on true contribution the moment it lands. This guide is for operators who want to catch loss making orders while they are still a handful rather than a month. It covers what contribution means here, which costs are real and which are estimated, and what to do when the radar flags something.
What contribution means
Contribution is what an order leaves behind after the costs that order caused. Revenue, less cost of goods, less payment processing, less shipping, less the advertising cost attributed to it.
It is deliberately not the same as gross margin. Gross margin treats every order in a product line as equivalent. Contribution recognises that two orders for the same product can differ enormously once you account for what was paid to acquire the customer and what it cost to deliver the box.
Operating expenses sit outside contribution, because they are not caused by any individual order. They are handled at the month level instead.
Real costs and estimated costs
This distinction matters more than anything else on the screen.
If ShipStation is connected, shipping is the actual label cost for that specific order. If it is not, shipping is your configured rate applied evenly, which is wrong at both ends: it overstates the cost of a small local order and understates a heavy interstate one.
If your cost of goods is matched at the product level, the figure is real. If it falls back to a blanket percentage, it is an assumption.
Tatheon prefers the actual over the estimate wherever it has one, per order and per day, rather than choosing one method for everything. So a month can contain both, and the closer your connections are to complete, the more of the radar is measurement rather than modelling.
What to look for
Individual loss making orders are normal and not always a problem. A first order acquired at a loss is how subscription and repeat purchase businesses work, and the question is whether the customer comes back, which is a retention question rather than a radar one.
The pattern is what matters. Look for clusters. A product that appears repeatedly is priced wrong or is heavier to ship than its price assumes. A region that appears repeatedly means your shipping rates do not reflect what that region costs. A discount code that appears repeatedly is doing more damage than the campaign report suggests, because the campaign report counts revenue and the radar counts what was left.
Acting on it
A clustered product usually needs a price change, a shipping rate change or a bundle that raises the order value enough to carry the freight. Removing the product is rarely the right first move, because a loss making product that brings in customers who then buy profitably is doing its job.
A clustered discount code usually needs a floor. Free shipping over a threshold that sits below your average freight cost quietly converts your best orders into your worst.
What it cannot tell you
The radar sees one order at a time. It does not know that this customer will order four more times, and it does not know that the loss making order came from a campaign that also produced twenty profitable ones. Use it to find patterns worth investigating, then use the cohort and retention sections to decide whether the pattern is actually costing you anything over a customer lifetime.