Online retail reports revenue and margin per product, both computed before returns. Returns then arrive as a single line in the accounts, at company level, so the cost never attaches to the product that caused it. A category returning at 8% and one at 45% look identical in a margin report.
The cost of a return is also badly understood. It is not the refund. It is outbound shipping already spent, return shipping, handling and inspection, repackaging, the markdown when an item cannot go back at full price, and the capital tied up while it travels. Most retailers count the refund and stop.
The finding that changes decisions is margin per product after all of it, because the ranking it produces is often the reverse of the revenue ranking. The product marketing pushes hardest is sometimes the one quietly consuming what the rest of the catalogue earns.
And return reasons are the cheapest product research available. "Too small" concentrated on one item is a sizing chart problem with a fixed cost to fix. "Not as described" is a photography and copy problem. Both are recorded and neither is read.
This agent reports true margin, the reasons behind it, and what each fix is worth. It changes no price, delists nothing, and never profiles a customer.
Which is what pushing a high-return product does.
A sizing chart correction, costed against what it recovers.
The one that should drive buying and marketing decisions.
Not the refund. Outbound shipping already spent, return freight, handling, markdown and tied capital.
Frequently the top seller sits near the bottom once returns are attached to it.
"Too small" on one item is a sizing chart. "Not as described" is photography. Both are already recorded.
What correcting one sizing chart is worth per month at current volume.
An item that comes back at full price is a different cost from one that cannot be resold at all.
It reports products and reasons, never individuals. Return-behaviour scoring of shoppers is a different and far more dangerous product.
Why E-commerce in particular. Margin reports are computed before returns, so a category returning at 8% and one at 45% look identical — and the top seller is sometimes the one consuming what the rest of the catalogue earns.
Runs unattended
Started by you or by an event, and it finishes on its own. Nothing waits for someone to be at a desk.
The same standard every time
The two-hundredth item is held to the bar the first one was. Consistency is the part people cannot sustain.
It cannot act on its own
Returns Margin Analyst has no path to sending, spending or committing. That limit is why its output is safe to act on.
This agent runs server-side through the PROMIVO runtime. Each run is logged step by step and every tool call is permission-checked before it executes.
Read-only by design. This agent has no path to sending, spending, publishing or committing anything. Where that limit is the product, removing it would remove the reason to trust the output.
Demo dataIllustrative sample output, abridged.
{
"period": "Q3 2026",
"currency": "USD",
"products": [],
"capitalCostPercent": 12
}{
"escalate": true,
"products": [
{
"sku": "SKU-2201",
"trend": "rising",
"marginRank": 94,
"returnRate": 44.1,
"revenueRank": 2,
"costOfReturns": 48000,
"marginAfterReturns": -6800,
"notResaleableShare": 18,
"marginBeforeReturns": 41200
}
],
"disclaimer": "An analysis of your own order and return records. No customer has been scored, profiled or identified, no price changed, no product delisted, no returns policy altered, and cost components you do not record are reported as missing rather than estimated.",
"reasonFindings": [
{
"sku": "SKU-2201",
"fixType": "sizing-chart",
"worthPerMonth": 11400,
"dominantReason": "Too small — 61% of returns on this item."
},
{
"sku": "SKU-1180",
"fixType": "photography",
"worthPerMonth": 3200,
"dominantReason": "Colour not as pictured — 48% of returns."
}
],
"escalationReason": "A top-decile revenue product has negative margin after returns, its return rate is 44.1% and rising across three periods.",
"rankingReversals": [
"SKU-2201 is your second best seller by revenue and 94th of 96 by margin after returns. Marketing has been scaling it all quarter."
],
"highReturnButHealthy": [
"SKU-3390 returns at 31% and still earns a healthy margin, because almost everything comes back resaleable at full price. Not every high return rate is a problem."
],
"costComponentsMissing": [
"Repackaging cost is not recorded, so every figure below understates the cost of a return."
]
}No integrations required.
True margin per product, after returns.
Whether the product being scaled actually earns.
Which lines to repeat, on margin rather than on revenue.
Which products and which reasons.
$299/month
Billed monthly through your PROMIVO subscription. Cancel at any time.
Runs consume your plan allowance for agent executions and tokens. See plan limits.
No, and it will not. Return-behaviour scoring of individuals is a different product with real fairness problems. This reports on products and reasons.
It uses the cost inputs you supply and states which components were missing, so the margin is understated in a stated way rather than estimated with an invented number.
No. Both are commercial decisions about range and positioning. It gives you the margin and the reasons.
No reviews yet. Reviews open once customers have run this agent.
Tell us what to change and our team will scope a customised version for your business.
Customize this agent