Mastering Profitability at scale on Amazon

A Brand can grow revenue and still end up worse off than when it started. It sounds like a contradiction, but it happens constantly in marketplace ecommerce, and it almost never shows up until the annual numbers are already in. Roughly one in three Amazon sellers lose margin as they scale, and fewer than 60% reach profitability in their first year. Growth and profitability are treated as the same conversation. They aren't, and the gap between them is where a lot of value quietly disappears.
Profitability lives at the SKU, not the P&L
The P&L shows whether the business made money last month or quarter. It doesn't show which products made it and which ones ate the profit of everything else. Averages hide that. A catalog can look healthy overall while half of it is subsidising the other half.
Getting to a true SKU-level number means walking down the full stack: selling price, minus Amazon fees, minus landed cost, minus advertising and promo, minus returns and refunds, minus losses and damages. What's left is contribution margin, and that's the number that ends up driving where inventory, ad spend and attention actually go, once a Brand starts looking at it directly.
There's a layer of cost that never touches a SKU
SKU-level margin is only part of the picture. A meaningful share of cost never attaches to any one product: overheads, agencies, accounting, compliance, legal, headcount, tools. This typically runs 15-25% of the business. Expansion costs sit separately again, legal entity and tax setup, supply chain, labelling, localisation, market-specific marketing, another 5-10% that's easy to underestimate.
Brands that only track SKU margin and leave this layer out tend to come away with a more flattering profitability number than the real one, sometimes flattering enough to justify a decision the underlying numbers wouldn't support.
Amazon's own reporting adds a layer of noise
Amazon's reporting is fragmented, and small errors compound. FBA and storage fees, lost and damaged stock, FX and cross-border fees, unsellable returns, fees that never get traced back to the SKU that caused them. Each one is small on its own. Stacked across a catalog, they can be the difference between a product that's profitable and one that isn't.
Checking the P&L often, rather than quarterly, and reconciling it line by line against what the reports actually say, is usually what catches this before it compounds.
Knowing the mix changes where resources go
Most catalogs we see follow something close to an 80/20 split: a handful of SKUs doing most of the work, the rest along for the ride. The mechanism behind this is simple: not every SKU converts a dollar of ad spend into the same amount of profit. A product with a 40% contribution margin returns far more from the same ad dollar than one sitting at 15%, even if the lower-margin product sells in similar volume.
Here's a simplified version of how that plays out: take a $100k ad budget split across a catalog where the average contribution margin is 25%, but the top SKUs run closer to 40%. Spread evenly across the catalog, that budget might produce $85k in contribution profit, a loss once ad spend is accounted for. Concentrated instead on the SKUs running at 40% margin, the same $100k could produce $110k in contribution profit, a modest profit after ad. Same money, same catalog, a $25k swing in outcome purely from shifting spend toward the higher-margin products. Spreading investment evenly feels like the fair thing to do. It is rarely the profitable one.
ROI as the deciding metric
Every growth decision, an ad channel, a promo, a supply chain switch, a new market or product, a new tool, can be evaluated on the same basis: whether the return justifies the cost. Ad spend scales when ROAS clears 4-5x. A promo makes sense when it covers the discount and the cannibalisation. A supply chain switch pays off when payback is under six months. A new market clears the bar when breakeven is under 12 months.
We ran that last test for a UK Brand in our portfolio deciding where to expand next: the US or the EU. The US felt like the obvious call, it's the largest ecommerce market by far. Once import duties, freight, and the marketing investment needed to get visibility in a market that size were priced in, the projected return came back at around -12%. The EU route looked less exciting on paper, but after assessing all costs and investment required, the ROI was roughly 10%. The bigger market lost on the numbers that actually mattered.
Infrastructure tends to match the stage
Below $500k in revenue, most Brands run on scrappy tools and manual tracking, with the priority on staying liquid. Between $500k and $5m, the shift is usually toward structure and tools built to scale. Above $5m, the focus moves to optimisation, integrated systems, and data-centric platforms.
The jump from 10 SKUs to over 100 tends to force this shift. Inventory management moves from manual reorder to automation. Competition goes from a handful of names to many, across markets. Cost structure becomes genuinely complex, and the impact of a single mistake stops being contained, it scales with the catalog. Complexity, in other words, grows faster than revenue does, which is the part most growth plans quietly leave out.
Why this matters now
Ten SKUs on one marketplace is a business that fits in a spreadsheet. Hundreds of SKUs across five channels and three markets usually doesn't, regardless of how well the team is running, and at some point the spreadsheet itself becomes the constraint rather than the tool.
That's roughly where tooling shifts from optional to necessary. We've spent most of the last several years building exactly that with the Brands we work with, inventory forecasting, pricing, cash flow visibility, marketing optimization, the kind of infrastructure that lets growth stay profitable instead of just getting larger. If any of this sounds close to where your own catalog sits, it's worth a conversation before the next growth push outpaces what's supporting it. Reach out to us to see where the margin is hiding in your business.
Related posts




.png)
.png)