The Margin report answers the question that revenue figures never do: after Amazon has taken its cut and you have paid for the goods, what is actually left?
It is one of the four views on the Analytics screen, and like the rest of Analytics it is available on Growth and above.
For every product, the report calculates:
margin = selling price minus Amazon fees minus your Shopify unit cost
That is the whole calculation. Its usefulness depends entirely on the quality of the three inputs, so it is worth knowing exactly where each one comes from.
The price your product is currently offered at on Amazon, in that marketplace's currency. If you are running the Repricer, this is the price the repricer last set, not your Shopify price. That is deliberate. The report tells you what you are earning today, not what you would earn at your list price.
The fees Amazon estimates it will charge on a sale at that price. Depending on the product and how it is fulfilled, this can include the referral fee, a closing fee, a per-item fee, and the FBA fulfilment, per-order and weight fees. The same breakdown appears on each product's detail page under Amazon fees, along with the net proceeds figure.
These fees are fetched from Amazon per listing, which means a product has to have been pushed to Amazon through Product Sync before the app has any fees for it. A product you have only mapped for price sync has no fee data.
This is the one people miss. The unit cost is the cost per item field you set on the Shopify variant, in Shopify itself. The app reads it. It never sets it, and it has no cost field of its own.
Warning
If you have not filled in the cost per item on your Shopify variants, the margin figures in this report will be wrong. With a blank cost, the calculation treats your goods as costing nothing, and every product will look far more profitable than it is. This is the single biggest cause of a margin report that reads too well.
Each row carries a margin badge, colour coded so you can scan the report rather than read it:
A Loss badge is not automatically a reason to pull a product, but it is always a reason to look. Sometimes it is a genuine loss maker. Often it is a data problem, which brings us to the caveats.
This report is an estimate, and it is honest about it.
Rows are flagged as estimates when an input is missing. If the app does not have fee data, a unit cost, or a current price for a product, the row is marked so you know the number in front of you was calculated with a gap in it. Take the flag seriously. A flagged row is not evidence of anything.
Amazon's fees are themselves estimates. They are Amazon's own projection of what it would charge on a sale at the current price. They are not a bill. Real charges can differ, for example if a product's weight band is reassessed or its referral category changes. Returns, storage fees and advertising costs are not in this calculation at all.
Two things fix almost every problem with this report.
Set your unit costs in Shopify. Open each product's variant in Shopify and fill in the cost per item. If you have many products, Shopify's bulk editor or a CSV import will do it far faster than clicking through them. Nothing here is trustworthy until this is done.
Push your listings through Product Sync. Fees are fetched per listing, so publishing a product is what makes its fee data appear. If a product shows no fees, check whether it is genuinely listed on Amazon or only price-mapped. The Listed on Amazon tile on the Dashboard tells you how many products are in the first category.
Once both are true, the flagged rows should largely disappear, and the ones that remain point at real gaps rather than missing homework.
Tip
Work the report from the bottom up. Sort by margin, look at every Loss badge and every red row, and check each one has a real cost set before you act on it. A single mistyped cost per item can make a healthy product look like it is haemorrhaging money.
Export CSV downloads the report as it appears, with the price, the fees, the cost and the calculated margin per row. Export it if you want to model a price change in a spreadsheet, or to add costs the report does not know about, such as inbound shipping, storage or advertising spend.