ACoS vs TACoS: Which One Should Actually Drive Your Decisions

Ask a seller how their advertising is doing and you'll usually get an ACoS. "We're at 24%." It's said the way you'd report a temperature, as though the number settles the question.

It doesn't, because ACoS answers a much narrower question than most people think it does. It's a good metric that gets asked to do a job it was never built for, and the gap between those two things is where a lot of PPC budget goes to die.

Here's what each metric actually measures, and what each one should be allowed to decide.

What ACoS measures

ACoS is ad spend divided by ad-attributed sales. If you spent $200 and Amazon credited $1,000 of sales to those ads, your ACoS is 20%.

The key word is attributed. Amazon draws a line between a click and a later purchase, and only sales inside that line count. Attribution windows differ by ad type, and they're not the same length, which is the first thing that makes cross-campaign comparison shakier than it looks. A shopper who clicks, thinks about it for a fortnight, and comes back may or may not be counted depending on which product of Amazon's you were running.

So ACoS is a campaign efficiency measure. Within one campaign, over one window, it tells you how much you paid for the sales the system was willing to credit to you. That is genuinely useful. It's how you tell a keyword that converts from one that burns money, and no other metric replaces it for that job.

What it isn't is a measure of what advertising did to your business.

The three things ACoS can't see

It can't see the organic sales your ads created. This is the big one. Advertising on Amazon isn't only a way to buy a sale, it's a way to buy rank. Ad-driven sales feed the sales velocity signals that move you up organic search results. Those later organic sales are real, they were caused by your spend, and ACoS assigns them a value of zero. A campaign at 45% ACoS that dragged a product from page four to page one has done something that a campaign at 18% ACoS on a term you already own has not.

It can't see cannibalisation. The mirror image of the same problem. If you're already ranked first organically for your brand name, ads on that term will post a beautiful ACoS. They're mostly buying sales you would have received for free. Low ACoS here is not efficiency, it's a discount you're paying yourself.

It has no denominator that means anything to the business. ACoS scales with the campaign, not with the company. You can hold a 20% ACoS while ad spend as a share of your revenue doubles, because the ad-attributed sales grew alongside it. Nothing in the number tells you that your business became more dependent on paid traffic.

What TACoS measures

TACoS is total ad spend divided by total sales, ads and organic together.

TACoS = total ad spend / total revenue

That's the whole formula. If you spent $4,000 on ads across everything and the account did $50,000 in sales, TACoS is 8%.

The change looks trivial. It isn't, because the denominator now includes every sale your ads helped cause but weren't credited for. TACoS asks a different question: what share of everything this business sells is going to Amazon for traffic?

Two notes on calculating it honestly.

Pull it per product, not just per account. Account-level TACoS is an average, and averages hide the product that's carrying a 30% TACoS inside a portfolio that reads 9%. The decisions you make are per product, so the number has to be too.

And be careful with the sales figure. It should be total sales for that product in that period, in the same currency, over the same date basis as the spend. Mismatching an order-date sales figure against a differently-bucketed spend figure produces a TACoS that wobbles for no reason and teaches you nothing.

What a healthy TACoS looks like

Any number quoted here is directional, not a benchmark. TACoS varies enormously by category, price point, competition and how old the listing is. Treat what follows as shape rather than target.

A new launch should have a high TACoS, and probably an uncomfortable one. You're buying rank you don't have. Spend is large relative to a small sales base, and most of the sales are paid. A TACoS in the tens of percent during launch is a cost of entry, not a failure, as long as it's temporary and you can fund it.

An established product should sit meaningfully lower. Organic has taken over a good share of the volume, ads are working the edges — defending your own terms, testing new ones, catching demand you don't rank for yet.

A mature brand with real organic strength runs lower still, and the interesting part is what the number is doing rather than where it sits.

The direction is the signal. As a product matures, TACoS should drift down while total sales hold or grow. That combination means advertising is compounding: each dollar you spend buys sales and buys rank, and the rank keeps selling after the dollar's gone. That is the entire thesis of Amazon advertising working properly.

If TACoS is flat while sales grow, ads and organic are growing in lockstep and you're renting your position rather than owning it. Not fatal. Worth knowing.

If TACoS climbs while sales are flat, you're paying more for the same business. Usually competition, sometimes a listing that stopped converting, occasionally a bid strategy quietly running away from you.

The trap: falling TACoS means two opposite things

This is the part that gets sellers, and it gets them in the direction of feeling good.

TACoS goes down when ad spend falls relative to total sales. There are two entirely different ways to produce that.

The good one. Organic sales grew. Your rank improved, the listing converts better, reviews accumulated, and a larger share of your volume now arrives without you paying for it. Spend stayed level, total sales rose, TACoS fell. This is what success looks like.

The bad one. You cut ads. Total sales haven't dropped yet, because organic rank has inertia — it reflects recent sales history, and that history still contains the period when you were spending. TACoS falls immediately. Sales fall later. You are coasting on rank you've stopped paying to maintain, and by the time the sales curve turns, you've lost position that costs more to win back than it would have cost to hold.

Both look identical on a TACoS chart. Same line, same direction, opposite futures.

Telling them apart takes about two minutes and one extra number: organic units, which is total units minus ad-attributed units.

• Organic units rising while TACoS falls: healthy. Your position is doing more work.
• Organic units flat or falling while TACoS falls: you're coasting. The metric improved because the numerator shrank, not because the business did.

Check organic sessions and conversion rate alongside it if you have them, and check your rank on the two or three terms that actually matter for the product. Rank decay is slow at first and then not slow at all.

The same logic runs in reverse, which is worth saying because it stops people panicking. TACoS rising during a deliberate launch push is expected. TACoS rising because you added a new product to the account is arithmetic, not a problem. Always ask which part of the fraction moved before you react to the fraction.

Which metric decides what

The clean split:

ACoS decides things inside the ad account. Which keywords to bid up or down. Which search terms to negate. Whether an ad group is worth the money. Which match type is doing the work. Anything where you're comparing two ad-level options against each other, ACoS is the right tool, because you're asking about relative efficiency within the same system.

TACoS decides things about the product. Whether the advertising strategy is working. Whether a launch is progressing or stalling. Whether this product deserves more budget or less. Whether you can afford to hold position through a competitive quarter. Anything where the question is "is this working for the business," TACoS is the number, because it's the only one whose denominator is the business.

And neither of them decides whether you're making money. That's a separate calculation involving your landed cost, Amazon's fees, returns and everything else that comes out of a unit. A product can run an excellent TACoS and still lose money on every sale. Efficiency and profitability are different questions, and running ads on the first without checking the second is how brands grow their way into trouble.

Which is the practical difficulty with all of this. TACoS is easy to define and tedious to maintain: you need total sales and total ad spend, per product, per marketplace, in one currency, on a consistent date basis, updated often enough to act on. Most sellers build it in a spreadsheet, run it monthly for a while, and stop.

That's the reporting Seller Sphere keeps standing for you, with organic units sitting next to TACoS so a falling line tells you which of the two stories you're in.