A good ACOS is any number below your break even ACOS, which is simply your margin. For most established sellers that means 15% to 25%, and our own lifetime average is 17.84%. Here is how to set your own target properly.
Amazon PPC
A good ACOS on Amazon is any number below your break even ACOS, which is simply your profit margin before advertising. Keep 30p of every pound after costs and you break even at 30%, so anything lower earns a profit on that advertised sale. For most established sellers that means working somewhere in the 15% to 25% range, and across the accounts we manage our lifetime average sits at 17.84%.
That is the short answer. The longer one matters, because the right number depends entirely on what you are trying to do, and anybody who quotes you a target without asking about your margin is guessing.
What is on this page
- What ACOS actually measures
- Break even ACOS, the only benchmark that is yours
- Why the target moves with the job
- A worked example, start to finish
- When a low ACOS is bad news
- The same product needs three different targets
- Category benchmarks, and why margin beats them
- How to improve ACOS without breaking anything
- The number ACOS cannot see
What ACOS actually measures
ACOS stands for advertising cost of sale, and it answers one question: for every pound of sales your ads produced, how much did you spend to get it. The maths is short. ACOS equals ad spend divided by ad sales. Spend £200 and those ads produce £1,000 of sales, and your ACOS is 20%.
A lower percentage means the advertising is more efficient. A higher percentage means each sale is costing more to win. What it does not tell you is whether those sales were incremental, which is the limitation we come back to at the end.
Break even ACOS, the only benchmark that is yours
Your break even ACOS is the point where the profit on a sale exactly cancels the cost of the ad that won it. It equals your margin before advertising, and it is the one number in this article that comes out of your own accounts rather than somebody else’s average.
Work it out once, write it down, and review it quarterly. Costs move, Amazon fees move, and a break even figure calculated against last year’s landed cost is not a benchmark any more, it is a habit.
Two things follow immediately. Anything below that line profits on the sale. Anything above it is buying something other than immediate profit, which can be entirely rational as long as you know what you are buying and for how long.
Why the target moves with the job
Launching. Forty to seventy percent is normal in the first three weeks, because that spend is buying data and the sales velocity that becomes organic rank. The right question is not whether the number is high, it is whether it is falling by month two. The full sequence is in the first 90 days.
Growing share. Running above break even on purpose, with a written end date, to take positions you intend to hold organically later.
Steady state. Where most established brands live, comfortably below break even, with the account maintained rather than pushed.
Maximising profit. Coasting on organic, paying only for the placements that genuinely add sales. Attractive, and it quietly hands ground to competitors if you stay there too long.
An account running all four jobs at a single target is managing none of them properly, which is why we set targets per campaign group rather than per account.
Not sure what your break even ACOS actually is?
Send us the reports and your costs. We will work out your real target and tell you plainly, free.
A worked example, start to finish
Take a product selling at £30 with a break even ACOS of 28%, as above. Last month it spent £1,400 on advertising and produced £5,000 of ad sales, so ACOS was 28%. Exactly break even, which sounds like failure.
Now add the rest of the picture. Total sales that month were £11,200, so total advertising cost of sale was 12.5%. Roughly £6,200 of revenue arrived without advertising paying for it, and a good share of that organic position exists because of paid sales in earlier months.
So is 28% good? On the campaign, it is break even and worth improving. On the business, the product is comfortably profitable and the advertising is doing its job of holding a position that produces free revenue. Two correct answers to two different questions, which is exactly why we never report one of these numbers without the other.
The improvement path here is not a bid change. It is finding the share of that £1,400 sitting on search terms that never convert, which in most accounts is a fifth to a third of the budget.
When a low ACOS is bad news
A 6% ACOS looks like a triumph on a report and is often a symptom. It usually means one of three things.
You are bidding almost entirely on your own brand terms, buying customers who typed your name and were coming anyway. You have cut so far that you only appear on the cheapest, lowest intent placements. Or you have stopped competing on the terms that grow the business, and a competitor is quietly taking them while your ratio looks excellent.
The test is simple: is total revenue growing? If ACOS is falling and total sales are flat or down, the efficiency is cosmetic. That relationship is exactly what ACOS vs TACOS exists to explain.
A number below your break even point is not automatically good. It only counts if the business behind it is still growing.
The same product needs three different targets
One product, three phases, three correct answers, and treating them as one is the most common reporting mistake we see.
Months one to three. The product has no rank, no reviews and no history. Every sale is doing double duty, closing revenue and building the velocity that earns organic position. Forty to seventy percent ACOS here is the cost of entry rather than a failure.
Months four to nine. Organic starts to carry part of the load, so the same campaigns produce a lower number without anybody changing a bid. This is the phase where you find out whether the launch actually worked, and the signal is the direction rather than the level.
Month ten onwards. The product should be sitting comfortably below break even, with advertising defending position and picking up demand organic does not reach. If it is still at launch levels after a year, the problem is rarely the campaigns.
Set the target per phase and write down when you expect it to change. An account where every product is measured against one account wide number will always look like it has three problems, when what it really has is three products at three stages.
Category benchmarks, and why margin beats them
People want a table of category averages. They are less useful than they look, because fee structures and margins differ so much within a category that the average describes almost nobody.
A 35% ACOS is comfortable on a premium product with 50% margins and ruinous on a consumable with 20%. Both might be in the same Amazon category. Use your own break even figure as the reference and treat published averages as background noise.
What does travel across categories is the shape of a healthy account: a mix of low ACOS brand defence, mid range core keywords carrying most of the volume, and a small allocation to higher ACOS discovery that feeds the other two.
How to improve ACOS without breaking anything
Cut the spend that never sells. Search terms with clicks and no orders past your break even click count. No rank value, no sales, so cutting them costs nothing. Method in the search term report guide.
Improve conversion rate. A listing that converts better lowers ACOS on every keyword at once, which no bid change can do. Start with the mobile checks.
Read your placements. Many accounts pay a premium for placements that convert no better than the free ones. The mechanics are in bid strategies and placements.
Reduce, do not slash. Ten to fifteen percent at a time on terms you already rank for, watching total sales. The safe sequence is in cutting ACOS without losing rank.
Want the waste found before you touch a single bid?
We mark up your own search term report with what we would cut first. Free, and yours to keep.
The number ACOS cannot see
ACOS only counts sales the ads were credited with. It cannot tell you whether those sales were incremental or whether the shopper would have found you anyway, and it has nothing at all to say about the organic half of your business.
That is why we report total advertising cost of sale alongside it on every account. ACOS is the steering wheel for campaigns. The total figure is the dashboard warning light for the business, and the two disagree often enough that reading only one is how brands end up optimising themselves into a smaller company with better ratios.
Want your targets set from your own margins rather than a benchmark?
Twenty minutes on a call and you will know what number to actually aim at.
Frequently asked questions
Anything below your break even ACOS, which equals your profit margin before advertising. Keep 30p of every pound after costs and you break even at 30%, so anything lower profits on that sale. For most established sellers the working range is 15% to 25%, and our lifetime average across the accounts we manage is 17.84%.
Take your selling price, subtract cost of goods, Amazon fees, fulfilment and a realistic allowance for returns. What is left, expressed as a percentage of the selling price, is your break even ACOS. Sell at £30 with £9 left after everything and you break even at 30%.
No. A very low ACOS often means you are only bidding on branded or bottom of funnel terms and leaving growth on the table while a competitor takes the positions you could have had. Low ACOS is good news when you are already winning the terms that matter.
Higher than your target, on purpose. Forty to seventy percent is common in the first three weeks because that spend is buying data and rank rather than immediate profit. What matters is whether it is falling by month two and three.
No, and that is its main limitation. ACOS only counts sales attributed to the ads. Total advertising cost of sale counts every sale you made, which is why the two together tell you far more than either alone.
Category matters less than margin. A 35% ACOS is comfortable in a category with 50% margins and ruinous in one with 20%. Fee structures differ enough between categories that the only benchmark worth using is your own break even figure.
No. Brand defence campaigns often run very low, discovery campaigns much higher, and launch campaigns higher still. A single account wide target flattens those differences and usually leads to cutting the campaigns doing the hardest work.
Send the right reports. Get a straight answer back.
We read the account properly and write back with where the money is going, what it is buying, and the three things we would change first. You keep it either way.
No obligation · no card · a person replies, usually the same working day