The short answer

ACOS — Advertising Cost of Sale — is the share of your attributed Amazon sales spent on advertising: ad spend ÷ attributed sales × 100%. Your break-even ACOS is the highest ACOS an order can carry before it loses money, set by your margin after Amazon's fees. When your actual ACOS sits below break-even, each ad-attributed order contributes profit.

Measure your actual ACOS

Enter two campaign totals from the same Amazon Ads report.

Ad spend$Total amount spent on ads Attributed ad sales$Sales attributed to those ads

ACOS—Enter your campaign totals

ROAS equivalent—Same return, expressed as ROAS

Calculation

Ad spend ÷ attributed sales × 100 = ACOS

Use attributed sales, not total store sales — mixing scopes breaks the measure. ACOS and ROAS are reciprocals when both use the same inputs; see ACOS vs ROAS.

Find your break-even ACOS

Enter your unit economics; use 0 where a cost does not apply.

Product price$Selling price per unit Product cost$Unit cost to make or source it Referral fee%Your category's Amazon referral rate Fulfillment / FBA$Per-unit fulfillment cost

Your actual ACOS (%) — optional, for the verdict%

Break-even ACOS—Enter your unit economics

Max ad spend per order—Contribution available for ads

Calculation

(Price − cost − fees) ÷ price × 100 = break-even ACOS

The referral fee is applied as a flat percentage of price. Amazon's real schedule varies by category and has per-item minimums — enter your category's effective rate.

Work through one product

Illustrative: a kitchen gadget sells at $29.99. The unit costs $8.50, the 15% referral fee comes to $4.50, and FBA fulfillment costs $5.20.

Non-ad costs: $8.50 + $4.50 + $5.20 = $18.20
Contribution before ads: $29.99 − $18.20 = $11.79
Break-even ACOS: $11.79 ÷ $29.99 × 100% = 39.32%

At a 20% actual ACOS, ads cost $29.99 × 20% = $6.00 per order, leaving $11.79 − $6.00 = $5.79 of contribution. At a 45% ACOS, ads cost $13.50 per order — about $1.70 more than the $11.79 available — so each ad-attributed order loses money. The verdict is always relative to your costs, never to a category average.

Know what this leaves out

Organic halo. ACOS counts only attributed sales. If ads lift your organic rank, the true return is higher than ACOS suggests — but don't guess the lift; measure it separately.
Returns. Amazon adjusts attributed sales when orders are refunded, so ACOS drifts after refund-heavy periods. Recompute rather than trusting a stale number.
Tax basis. Keep the price tax-inclusive or tax-exclusive on both sides of the break-even formula; mixing bases moves the answer.
Fee minimums. A flat referral percentage ignores category minimums and tier breaks. Close enough for planning, not for accounting.
Attribution window. Sales outside Amazon's attribution window (commonly 14 days) never enter the denominator, so slow-converting products read worse than they are.

Keep these

Gather: take ad spend and attributed sales from the same Amazon Ads report and the same date range.
Produce: actual ACOS = spend ÷ sales × 100%.
Compare: break-even ACOS = (price − cost − fees) ÷ price × 100%. Scale spend only while actual stays below it.
Revisit: recompute when the price, fee schedule, or refund rate changes.

Decision boundary: actual ACOS above break-even means ad-attributed orders lose money. Fix the inputs — lower bids, raise the price, or cut costs — before raising spend.

Next reading

Examples are illustrative, not client results or benchmarks. The calculations use the inputs shown; real-world decisions depend on your records and assumptions. Report a correction.