Most Amazon advertising reviews open with a ratio and close with a ratio. ACoS came down, so the quarter went well. TACoS came down, so the brand is getting healthier. Both of those sentences can be true in a quarter where the business made less money than the one before it, and on a large catalog they usually are.
The problem is not that the two metrics are wrong. They are both doing exactly what they were designed to do. The problem is that neither of them is a profit number, and they are almost always presented as though they were.
What Each Number Actually Measures
Strip the reporting away and the definitions are simple. ACoS divides advertising spend by the sales that advertising was credited with. TACoS divides the same advertising spend by total sales, advertised and organic together.
| Metric | Spend divided by | The question it answers |
|---|---|---|
| ACoS | Ad-attributed sales | How efficient is the advertising, inside the advertising? |
| TACoS | Total sales | How dependent is the business on advertising? |
| Contribution margin | Nothing. It is a subtraction, not a ratio. | Did this product make money? |
Look at what is missing from the first two rows. Referral fees are not in them. Fulfilment is not in them. Storage, returns, removals and cost of goods are not in them. Both ratios can improve while every one of those lines gets worse.
ACoS Falls for Three Reasons and Only One of Them Is Good
When ACoS improves, the reporting deck treats it as a single event. It is not. There are three distinct causes and they have opposite implications.
- Bids and targeting genuinely improved. The same demand was captured for less money. This is the outcome everyone assumes they are looking at, and it is the least common of the three.
- Spend was cut on the terms that were feeding organic rank. The ratio improves immediately because the denominator holds for a few weeks on existing momentum. Organic position decays with a lag, so the damage lands one or two reporting cycles after the improvement was celebrated.
- Price went up. A higher price raises ad-attributed sales per conversion, so the ratio falls without anything about the advertising changing. Unit volume may be falling at the same time.
Any advertising review that reports a change in ACoS without saying which of those three produced it has reported a number, not a finding.
TACoS Is the Better Metric and Still Not a P&L Number
TACoS is a real improvement on ACoS because it cannot be gamed by shifting sales between advertised and organic. If TACoS is falling while revenue holds, the brand is genuinely earning more of its demand rather than renting it. That is the direction you want.
TACoS falling tells you advertising dependence is dropping. It does not tell you the drop was free.
Total sales sit in the denominator, so anything that lifts total sales improves TACoS. A deep coupon improves TACoS. A Lightning Deal improves TACoS. A price cut improves TACoS. In each case sales rise, advertising spend holds, the ratio falls, and contribution margin per unit falls with it. The metric moves the right way for the wrong reason, and nothing in the advertising console will tell you that happened.
The Number Underneath Both
Contribution margin per unit is the layer where advertising decisions and catalog decisions finally meet. It is a subtraction, and it is worth writing out in full because the order matters:
Selling price, less the referral fee, less fulfilment, less a returns provision, less storage, less cost of goods, less the advertising allocated to that unit. What is left is what the unit contributed.
Two things about that line are worth stating plainly. First, advertising is the last subtraction, not the first, so a SKU can be advertising-efficient and still lose money on every unit. Second, the returns provision belongs in the per-unit maths and is routinely left out, because returns arrive weeks after the sale and land in a different report.
Where the Three Belong in a Real Review
These are not competing metrics. They are three different altitudes, and the mistake is reading one at the altitude of another.
- ACoS is tactical. Campaign and target level, read weekly, used to decide bids and negatives. It answers a question about the advertising and should never be asked a question about the business.
- TACoS is directional. Brand level, read monthly, used to see whether organic demand is compounding or eroding.
- Contribution margin is the decision layer. SKU level, read monthly, used to decide what to advertise, what to reprice, what to repackage, and what to stop selling.
Takeaway
Keep both ratios. Add the subtraction. A brand that reviews ACoS weekly, TACoS monthly, and contribution margin per SKU monthly will make different decisions from one that reviews only the first two, and the difference will show up in the bank rather than in the deck.
Three Questions for Your Next Advertising Review
- ACoS improved. Which of the three causes was it, and what is the evidence?
- TACoS improved. Did organic grow, or did we discount our way to a bigger denominator?
- Which SKUs are we currently advertising into negative contribution, and are we doing it deliberately?
The third question is the one that changes the meeting. In most catalogs the answer exists, nobody has been asked for it, and the SKUs involved are not the ones anyone would have guessed.