For most of our time on Amazon, the number our ad reports led with was ACoS. Over the last year it drifted from about 40% to 59% while our ad-attributed sales stayed flat at around $20,000 a month. Was that bad? ACoS alone couldn't tell us. That's the problem with it.
ACoS and TACoS in one line each
- ACoS (advertising cost of sale) = ad spend divided by the sales Amazon attributes to your ads.
- TACoS (total advertising cost of sale) = ad spend divided by all your sales on Amazon, ads and organic together.
ACoS grades your ads on their own. TACoS asks whether your ads are growing the whole business.
Break-even ACoS: the number every SKU needs
Before you can say an ACoS is good or bad, you need to know the ACoS at which an ad sale makes you no money. That's your break-even ACoS, and it's simply your profit margin before advertising:
Break-even ACoS = (price − product cost − Amazon fees − shipping) ÷ price
An example with round numbers: a $50 product that costs $12 to make, with $16 in Amazon referral and FBA fees, leaves $22 before ads. Break-even ACoS is 22 ÷ 50 = 44%. Spend more than 44 cents per ad dollar of sales and that sale loses money.
Do this for every product. Ours range a lot, because a 30-day multivitamin pack and a 24-count pack don't have the same margin.
Why one blanket ACoS target fails
Most ad setups run one target ACoS across the whole account. If your products have different margins, that single number is too strict for your high-margin products and too loose for your low-margin ones. You underspend where you could grow and overspend where you lose money, and the blended number looks fine the whole time.
Set a target per product from its own break-even, not one number for the account.
What TACoS shows that ACoS hides
Ads do more than produce the sales Amazon credits to them:
- Organic lift. Ad sales help your product rank, which brings in sales nobody clicked an ad for.
- Repeat buyers. For a consumable, the first order is often bought with an ad and the next five aren't. Subscribe & Save makes this even stronger.
- Branded searches. People who saw your ad last week search your brand name this week.
If your ACoS rises but TACoS falls, your ads are feeding organic growth, which can be a good trade. If both rise together, as our ACoS did while sales stayed flat, you're paying more for the same business.
What's a good TACoS?
There's no universal number, but there is a pattern. A brand launching or pushing for rank can run a high TACoS on purpose. A mature brand should see TACoS come down over time as organic and repeat sales grow. Watch the direction month to month rather than chasing a benchmark.
How to lower ACoS without starving growth
- Cut the zero-order search terms. This is the fastest win. See our negative keywords guide.
- Stop paying twice. When several of your campaigns serve the same search, the most aggressive bid wins.
- Split brand from generic. Branded searches convert at a much higher rate. In our account, branded terms returned about $8 for every $1 spent while generic terms returned about $1. A blended number hides both.
- Bid by the hour. Paying the same at 3am as at 7pm wastes money. More in our dayparting post.
Setting targets by product margin
Put your product costs next to your ad data and give each product its own target: below break-even for steady profit, near break-even for products you're pushing, and above it only on purpose for a launch. Then track TACoS monthly to make sure the whole business is moving the right way.
That's how Adtender sets bids: from each product's margin and your TACoS, not from one account-wide ACoS.
Bids set from your margins
Adtender is in private beta, running on our own brand first.
Request early access