TACOS, Not ACoS: The Measurement Shift

Most Amazon sellers can recite their ACoS to the decimal point and have never once calculated their TACOS. That's backwards. ACoS tells you whether a single campaign is efficient. TACOS tells you whether your business is becoming less dependent on paid traffic over time — which is the actual question that determines whether you have a durable listing or a rented one.

TACOS

Total Advertising Cost of Sale — total ad spend divided by total sales (organic plus advertising), expressed as a percentage. It measures how much of your entire revenue is being paid for with ads, not just the revenue ads directly touched.

What is TACOS on Amazon?

TACOS answers a simple question: of everything I sold, what share did I pay for with ads? You calculate it as total ad spend across all campaigns, divided by total sales across the whole listing or account — organic sales, ad-attributed sales, and (increasingly) sales that started as a Rufus or AI-surfaced recommendation, all summed together.

That denominator is the entire point. ACoS only looks at the slice of revenue Amazon attributes directly to an ad click. TACOS looks at everything. A seller obsessed with ACoS can optimize a campaign into a beautiful number while the business underneath it barely grows, because the campaign is mostly buying sales that would have happened anyway. TACOS is the metric that would catch that — ACoS, by design, cannot.

TACOS vs ACoS: what is the difference?

ACoS (Advertising Cost of Sale) is ad spend divided by ad-attributed sales — the standard campaign-level efficiency metric every seller already watches in Seller Central. TACOS is ad spend divided by total sales. Same numerator, completely different denominator, and that difference changes what each metric can and can't tell you.

Here's the failure mode ACoS alone invites: a seller sees a 15% ACoS on a campaign, calls it efficient, and scales the budget. Total revenue climbs. Everyone's happy. But if organic sales for that same SKU quietly fell over the same period, the business didn't grow — it substituted paid traffic for the free traffic it used to get. ACoS never shows you that trade because organic sales aren't in its formula at all. TACOS is.

A worked example

Take a listing doing $50,000 in a month: $20,000 from ad-attributed sales and $30,000 organic. Ad spend for the month is $3,000.

• ACoS = $3,000 ÷ $20,000 = 15%. Read alone, that looks like a well-run campaign.
• TACOS = $3,000 ÷ $50,000 (total sales) = 6%. This is the healthier number to anchor on — it says ads cost 6 cents of every revenue dollar across the whole listing.

Now run the same listing three months later. Ad spend has doubled to $6,000 and ad-attributed sales grew proportionally to $40,000, holding ACoS steady at 15%. Total sales, though, only moved from $50,000 to $58,000 — because organic sales quietly dropped from $30,000 to $18,000.

• ACoS = $6,000 ÷ $40,000 = 15%, unchanged. By this metric alone, nothing looks wrong.
• TACOS = $6,000 ÷ $58,000 = ~10.3%, up from 6%. This is the number that flags the problem: ad spend is doing more and more of the work, and organic demand — the free traffic — is shrinking underneath the paid growth.

Why ACoS alone hides ad cannibalization

Cannibalization happens when an ad serves a shopper who would have found and bought the product anyway — through organic search, a repeat-purchase habit, or increasingly, a direct recommendation from Rufus or an off-site AI assistant. Amazon's attribution model counts that sale as ad-driven the instant a click happens in the path, even if the click added nothing. ACoS has no way to separate a converted-anyway sale from a genuinely incremental one, because both land in the same numerator.

The sellers most exposed to this are the ones with strong catalogs — complete backend attributes, solid A+ content, healthy review volume. Ironically, a well-optimized listing is more likely to get cannibalized by its own ads, because it was already going to convert organically. See Amazon listing optimization for Rufus for what that catalog work looks like. If you've done it well, watching TACOS is how you find out whether your ad spend is still earning its keep or just paying rent on demand you already had.

What is a good TACOS for Amazon sellers?

There is no single good TACOS number, and treat anyone who quotes you one flatly as guessing. A new listing with no review history and no organic rank often runs a TACOS well above 20% for months — that's expected; you're buying the organic rank and reviews that will eventually carry the listing. A mature, well-reviewed hero SKU in a low-competition category might sit comfortably under 5%. Category competitiveness, price point, and listing age all move the baseline.

What matters is the direction of the trend, tracked monthly or quarterly, against total revenue:

1. 1

   Falling TACOS, flat or growing revenue

   The healthiest pattern. Organic and Rufus-surfaced demand are carrying more of the sales, and ad spend is filling in the gaps rather than doing the heavy lifting. This is what a maturing, catalog-first listing should look like.

2. 2

   Flat TACOS, growing revenue

   Acceptable and common during active scaling — you're growing the pie and ad spend is growing proportionally with it. Watch it, but it's not a red flag on its own.

3. 3

   Rising TACOS, flat revenue

   The warning sign from the worked example above. You're spending more to hold the same ground, which usually means organic or Rufus-driven demand is eroding underneath the paid layer.

4. 4

   Rising TACOS, rising revenue

   Ambiguous — could be healthy launch-phase investment or an early warning depending on how much of that revenue growth is genuinely incremental. Cross-check against organic sales share specifically, not just total revenue.

Reading TACOS alongside Rufus-surfacing signals

TACOS tells you that ad dependency is falling. It doesn't tell you why — specifically, it can't distinguish organic search rank improving from AI-driven discovery picking up the slack. That distinction matters more every quarter, because Rufus and off-site assistants like ChatGPT, Claude, Gemini, and Perplexity are an increasingly large and untracked share of that organic-looking demand.

For catalog-first sellers, the useful practice is to read TACOS next to a citation-style metric: how often your product actually gets surfaced when a shopper asks Rufus or an AI assistant a category question, independent of any ad. There's no official dashboard for this yet, so it takes deliberate checking — running the questions your category gets asked and noting whether your product comes up. Structured, complete backend attributes are what make that surfacing possible in the first place; see Amazon backend attributes and the Rufus test for your listing for how to check it directly.

• Falling TACOS + rising Rufus-surfacing — the strongest signal. Your catalog work is compounding: AI discovery is replacing ad spend, not just organic search rank.
• Falling TACOS + flat Rufus-surfacing — likely organic search rank improving on its own; still healthy, just a different mechanism.
• Flat or rising TACOS + flat Rufus-surfacing — no compounding effect yet. Ad spend is still the load-bearing channel, which is fine early but worth investing catalog effort against.

How to start tracking TACOS this week

You don't need new software to start. Amazon's Business Reports give you total ordered product sales by ASIN and period; your ad console gives you total ad spend for the same period. Divide one by the other.

1. 1

   Pull total sales and total ad spend monthly, per hero SKU

   Use Business Reports for total sales (not just ad-attributed) and Campaign Manager or Seller Central advertising reports for spend. Match the same date range for both.

2. 2

   Calculate TACOS = total ad spend ÷ total sales

   Do this per SKU, not just at the account level — account-level TACOS averages away exactly the per-listing warning signs you're trying to catch.

3. 3

   Log it alongside ACoS, not instead of it

   You still need ACoS to manage individual campaign efficiency. TACOS answers a different question and neither one replaces the other.

4. 4

   Review the trend quarterly, and after any catalog change

   TACOS moves slowly and noisily month to month. Judge it on a rolling trend, and specifically re-check it after enrichment work, new A+ content, or a listing overhaul to see whether organic and AI-driven demand responded.