Here's a conversation we have in almost every first audit. The brand shows us their dashboard. Account ACoS: 24%. Perfectly respectable. Everyone's been reporting it monthly for two years. Then we rebuild the numbers per product, and the healthy average turns out to be two excellent products carrying four mediocre ones and two that have been losing money on every advertised unit for eighteen months.
Nobody did anything wrong, exactly. They just measured at the wrong altitude. This piece is the case for measuring advertising the way we do it at rankhouse: TACoS, per ASIN, against each product's own breakeven. It's more work. It's also the difference between an ad account that generates revenue and one that generates profit.
Why ACoS flatters and TACoS tells the truth
ACoS divides ad spend by ad-attributed sales. It answers a narrow question: how efficient were the clicks I paid for? It says nothing about the organic sales your advertising supports, and nothing about the organic sales it cannibalises. A campaign can hold a beautiful ACoS while your total sales flatline, because it's increasingly paying for orders you would have won organically.
TACoS divides ad spend by total sales, organic included. It answers the question that actually matters: how dependent is this product's revenue on paid traffic? A falling TACoS with stable spend means organic strength is compounding, which is the whole point of advertising on Amazon. A rising TACoS means you're renting a bigger share of your own revenue every month.
| Blended account ACoS | Per-ASIN TACoS | |
|---|---|---|
| What it measures | Ad spend against ad-attributed sales, averaged over everything | Ad spend against total sales, product by product |
| What it hides | Loss-making products inside a healthy-looking average | Very little; that is the point |
| The decision it supports | Roughly none beyond "spend more or less" | Bid ceilings, kill lists, scaling the winners |
| Failure mode | Same revenue, quietly shrinking margin | Requires per-product cost data to be accurate |
Why "per ASIN" is not optional
Every product in your catalogue has a different gross margin, a different fee stack, a different competitive set and a different price elasticity. Which means every product has a different breakeven TACoS: the advertising share of revenue at which a unit stops making money. A £45 supplement with a 60% margin before ads can sustain a TACoS that would bankrupt a £14 product with a 35% margin.
Blend those products into one account number and you've built an instrument that cannot detect the exact failure it exists to catch. The strong product's headroom masks the weak product's losses. The average stays green while individual products quietly die inside it. This is not a hypothetical: in one anonymised account from our own book, a haircare brand, restructuring around per-product economics took TACoS from 31% to 9% at the same revenue, and profit roughly tripled. Nothing about the products changed. The altitude of the measurement did, and the spending decisions followed.
How to build it, practically
- Cost every ASIN to true profit first. Landed cost, referral fee, fulfilment fee, storage share, returns rate, EPR and packaging costs. This gives you margin before ads, per product. Painful once, invaluable forever.
- Derive breakeven TACoS per product. If margin before ads is 40% of price, then 40% is the TACoS at which advertising eats the entire unit economics. Your working ceiling sits well below it, wherever you need profit to land.
- Report TACoS per ASIN weekly, against its own ceiling. Not monthly, and never blended. Weekly is the cadence at which drift gets caught before it compounds. This is exactly how the rankhouse client dashboard is built: 26+ metrics per ASIN, EBITDA per unit, TACoS per product, never blended.
- Let the number drive one of three actions. Under ceiling and falling: scale. Near ceiling and stable: hold and work the organic levers. Over ceiling: cut spend, fix the listing, or accept the product is being kept alive artificially and decide its future on purpose.
Want this level of thinking inside your account?
rankhouse is a boutique specialist agency: experienced Amazon professionals, no juniors, led personally by founder Dan Whalley. Fixed fee from £1,500 + VAT a month. Every engagement starts with a free profit audit built from your own data.
Get the free audit at rankhouse.co.uk →The objections, answered
"TACoS punishes launches." True, and correctly so. A launch runs high TACoS deliberately, buying rank and reviews. The discipline isn't a low number from day one; it's a planned glide path. High TACoS with a destination is investment. High TACoS without one is a subsidy, and the difference should be written down before the first pound is spent.
"Our agency reports ROAS." ROAS is ACoS upside down and inherits every blind spot, with an extra one: it sounds like profit while measuring revenue. A 5x ROAS on a product with a 15% margin is a machine for converting your cash into Amazon's. We've said it before: vanity ROAS is a tax on growth.
"This is a lot of spreadsheet." It is, the first time. That's why most agencies don't do it: it's harder to produce and harder to hide behind. But once the per-product model exists, it updates weekly with an hour of work, and every advertising decision for the rest of the account's life gets made against real economics. The related discipline of hunting spend that produces nothing at all is covered in the kill list.
The test to run this week
Take your five biggest products by spend. For each, pull total sales and total ad spend for the last 90 days, and calculate TACoS against a margin-based ceiling. If you can't complete the exercise because the per-product margin data doesn't exist, that finding matters more than any number would have: it means every advertising decision in the account is currently being made blind.
That exercise, run across a whole catalogue, is the core of the free audit we offer at rankhouse. We show you which products are funding the account, which are coasting, and which are being kept alive by budget that belongs elsewhere. Brands are sometimes surprised. The numbers never are.
Questions we get asked about this
What's a good TACoS number for an Amazon brand?
The honest answer is that the question contains the mistake: there's no universal good number, because TACoS is only meaningful against a specific product's margin structure. A product with 60% margin before ads can run 20% TACoS and bank healthy profit; a product with 30% margin at the same TACoS is donating a third of its economics to the auction. The useful question is per product: what's this ASIN's breakeven TACoS, derived from its true margin, and where does it currently sit against that ceiling and its own trend? A falling TACoS with stable spend signals compounding organic strength. A rising one signals growing rent. Both matter infinitely more than any benchmark.
How is TACoS different from ROAS, practically?
ROAS divides ad revenue by ad spend, which makes it ACoS upside down with an extra hazard: it sounds like profitability while measuring revenue. A 5x ROAS on a 15% margin product loses money on every advertised unit, and the dashboard applauds throughout. TACoS divides ad spend by total revenue, organic included, which is why it catches the two failure modes ROAS structurally cannot: paid clicks cannibalising organic orders you'd have won anyway, and products whose entire revenue base is becoming ad-dependent. Agencies report ROAS because it produces the biggest number with the least accountability. We report TACoS per product against breakeven because it's the only version that maps to money.
Isn't calculating breakeven per ASIN a huge amount of work?
The first pass is genuinely effortful: landed costs, the full Amazon fee stack, returns rates and packaging costs per product, assembled once. That's also precisely why it's rare, and why doing it is an advantage rather than a chore. Once built, the model updates weekly in under an hour, fees refresh from reports, and every advertising decision for the life of the account gets made against real economics instead of vibes. Set the effort against the alternative: an account spending five or six figures annually on ads with no product-level definition of success, which is the actual state of most accounts we audit. The spreadsheet is cheaper.
What do I do with a product that's over its breakeven TACoS?
Diagnose before cutting, because a breached ceiling has three different causes with three different treatments. If spend is leaking to irrelevant or zero-sale terms, it's a targeting problem: run the kill list and restructure. If clicks are relevant but the page doesn't convert, it's a listing problem, main image, price position, review depth, and no bid change fixes it. If the product converts fine but the margin simply can't carry competitive click costs, it's a product economics problem, and the honest options are price, cost reduction, or accepting a deliberate strategic subsidy with a written end date. What's not an option is the default: leaving it, because the blend hides it.
How often should TACoS be reviewed, realistically?
Weekly, per product, and the cadence is the discipline. Monthly reviews mean a drifting product burns four to five weeks of budget before anyone notices; quarterly means a season. Weekly catches the drift while it's a correction rather than a post-mortem, and it's frequent enough to read the direction of travel, which matters more than any single reading. This is exactly why the rankhouse dashboard reports 26+ metrics per ASIN every Monday with TACoS per product against its own breakeven, never blended: not because more data is virtuous, but because weekly resolution at product level is the minimum altitude at which an ad account can actually be steered.
The next step is twenty minutes.
If any of this reads like your account, the fastest way to find out is the free audit: per-product profitability, where the ad spend is leaking, and what we would fix first. No pitch deck, no obligation.
Book a chat with Dan →
About the author
Daniel Whalley is the founder of rankhouse, a boutique specialist agency for Amazon-focused growth in FMCG, health, wellness and beauty brands. He has spent 10 years inside Amazon accounts, generating £100M+ for the brands he works with, and manages £500k+ a month in ad spend across the UK, Europe and the US. He writes from inside the accounts he runs, not from the sidelines. Connect on LinkedIn → · amazon@rankhouse.co.uk