Published: 3rd September 2026
A rising Amazon sales graph can hide a deteriorating business. If advertising is doing more of the work each month, fees are climbing, discounts are becoming routine and inventory is sitting longer, top-line growth is not the result to celebrate. Amazon profitability analysis shows whether each pound of revenue is producing cash, or simply creating more operational pressure.
For established GB brands, the question is rarely whether Amazon can grow. The question is whether it can grow without consuming margin, working capital and management time. That requires more than checking ACoS in Advertising Console or looking at a monthly settlement report.
Amazon reports sales clearly. Profit needs to be constructed.
A product can appear successful because it has strong conversion, a prominent organic rank and year-on-year sales growth. Yet it may be unprofitable once referral fees, FBA fulfilment fees, inbound freight, returns, promotional discounts, VAT treatment, storage and advertising are included. This is particularly common in categories where price competition is aggressive and sponsored placements are costly.
The practical unit of analysis is the SKU or ASIN, not the account average. Account-level metrics blend winners with products that are quietly draining margin. They can also disguise an advertising programme where a handful of hero products generate genuine contribution while the rest rely on branded search or defensive spend to survive.
Start with net sales, not gross sales. Then deduct the costs directly attached to generating and fulfilling those sales. What remains is contribution before central overheads. This is the number that tells you whether increased Amazon volume deserves more inventory and more budget.
A useful Amazon profitability analysis brings commercial, operational and advertising data into one view. The aim is not a perfect finance model on day one. The aim is a decision-grade view of where money is made, where it is lost and what should change first.
For every meaningful ASIN, calculate:
Do not use only the current purchase price from a supplier. If stock was bought under a different exchange rate, shipped at a higher freight cost or carries duty that has not been allocated properly, the margin model is fiction. Use landed costs that reflect the inventory being sold now, and update them whenever purchasing conditions move.
The same discipline applies to returns. A return rate of 8% has a very different effect on a low-priced FBA item than on a premium product with strong gross margin. Returned units may be resellable, but they still generate fulfilment, processing and customer-service costs. Treating returns as a footnote is how seemingly healthy ASINs become margin traps.
ACoS is a useful delivery metric. It is not a profitability metric.
An ACoS of 25% can be excellent for an ASIN with 55% contribution before advertising and impossible for one with 22%. The right target depends on price, fees, landed cost, return rate, organic halo and the role that ASIN plays in the wider range.
This is why a single account-wide target ACoS creates poor decisions. It can force profitable growth products to underinvest while allowing low-margin products to keep spending because their reported ACoS looks acceptable.
Set a break-even ACoS at ASIN level. In simple terms, this is the maximum percentage of attributable sales that can be spent on advertising before the product stops contributing profit. Then set a trading target below it, leaving room for the profit the business expects to retain.
For example, if an ASIN leaves 38% contribution before advertising, its theoretical break-even ACoS is 38%. That does not mean a 37% ACoS is a good outcome. It merely means the sale has not yet lost money on direct costs. A target of 20-25% may be more appropriate if the brand needs cash to fund stock, payroll and growth.
TACoS adds necessary context because it compares total advertising spend with total sales. If TACoS rises while organic sales remain flat, paid activity may be replacing demand you previously earned organically. If TACoS remains stable as sales scale, advertising may be supporting genuine growth. Neither metric should be read in isolation.
Not all Amazon advertising should be expected to deliver the same immediate return.
Defending branded terms may protect conversion and visibility against competitors. Launching a new variation can require investment before ranking and review volume develop. Expanding into a new generic search term can be strategically correct even when its early ACoS is above the mature account target.
The failure is not spending money on these activities. The failure is mixing them into one budget and calling the total “performance”.
Create distinct budgets for proven profit, controlled growth and defence. Proven-profit campaigns should have clear contribution targets and reliable pacing. Growth campaigns need a fixed test window, a defined spend limit and a decision point. Defence should be monitored for competitor pressure, branded conversion and the cost of retaining demand you already created.
This structure stops experimental spend quietly becoming permanent. It also gives founders and finance teams a more honest explanation of why the advertising bill is changing.
The largest margin opportunities are often not hidden in a dramatic spreadsheet error. They sit in small, repeated inefficiencies across the catalogue.
Search term waste is one example. Broad and automatic campaigns can reveal valuable terms, but they can also keep funding irrelevant queries, poor-converting product targets and duplicate traffic. Review spend at search-term level, not just campaign level. Cut terms that have passed a sensible click or spend threshold without a profitable return, then move proven terms into tightly managed exact-match campaigns.
Retail readiness is another. A weak main image, unclear pack size, thin comparison content or a price that does not match the perceived value will depress conversion. Paid traffic then becomes more expensive because the detail page cannot turn visits into orders. Advertising optimisation and listing optimisation are not separate workstreams when profitability is the objective.
Inventory creates a third leak. Slow stock accrues storage charges and ties up cash. Stock-outs damage rank and force expensive recovery spend when inventory returns. Review contribution alongside weeks of cover, aged inventory and replenishment lead times. A high-margin ASIN that repeatedly goes out of stock is not being managed for profit.
Percentages make performance easy to compare, but pounds determine what the business can reinvest.
A campaign with a 15% ACoS may generate only £200 in contribution each month. Another at 28% may generate £4,000 after ad spend because the average order value and pre-ad margin are stronger. The second campaign may deserve more budget even though its headline efficiency is lower.
Use a weekly trading view for advertising decisions and a monthly view for full profitability. Weekly reviews should focus on spend pacing, conversion shifts, search term waste, stock risk and material changes in contribution. Monthly reviews should incorporate returns, fees, promotions and inventory costs that may lag behind the click.
That cadence matters. Overreacting to two days of data leads to unnecessary bid changes. Waiting for a quarter means waste becomes embedded. The right frequency depends on volume, but the principle is consistent: move quickly enough to protect margin, slowly enough to avoid noise-led decisions.
The output of an Amazon profitability analysis should not be a spreadsheet that confirms everyone is busy. It should produce a ranked action plan.
Classify ASINs into four commercial groups: scale, fix, defend and exit. Scale products have healthy contribution, stable conversion and room for additional demand. Fix products have clear potential but a solvable issue, such as poor advertising structure, weak conversion or an unallocated fee problem. Defend products protect a strategically important position but require disciplined spend. Exit products cannot reach an acceptable contribution level without a material change to price, cost or fulfilment.
This classification forces decisions. Some products need higher prices, even if conversion falls slightly. Some require supplier renegotiation or a pack-size change. Some need advertising cut immediately. Others deserve more stock and a larger share of budget because the economics prove they can scale.
Accendo360 approaches this work as a commercial control system, not a PPC reporting exercise: establish the margin truth, rebuild campaign architecture around it, then manage spend against a clear profit objective.
The useful question for your next Amazon review is not, “Did sales grow?” Ask which ASINs generated contribution after every variable cost, which ones consumed it, and what you will change before the next purchase order is placed. That is where profitable scale starts.