Your break-even ACoS is about half what you think it is

Every break-even ACoS calculator on the internet gives you the same answer. Break-even ACoS equals your profit margin before advertising. Margin is 30 percent, break-even is 30 percent, run under it and you make money.

That is correct arithmetic for a market where almost every order stays sold. It is wrong for India, and it is wrong in the direction that costs you money rather than the direction that costs you sales.

Here is the version with returns in it.

First, two words that get used interchangeably and should not be

RTO is return to origin. The order never reached the buyer. Refused at the door, address unreachable, or cash on delivery and nobody came out. You paid to ship it out and you pay to ship it back, and you never had a customer.

Customer returns are orders that were delivered, kept for a while, and sent back.

Both of them undo the sale. Both of them cost you logistics twice. Depending on your category, Indian marketplace sellers commonly see the two together eating 20 to 30 percent of orders, with apparel and fashion at the punishing end of that and small electronics and consumables much lower.

For the rest of this I am going to use one combined rate and call it returns, because for the arithmetic it does not matter which of the two took the order away from you.

The number your calculator gives you

Take an illustrative product.

So you set a target of 25 percent, feel disciplined about it, and go to work.

The number with returns in it

Now run one hundred orders through the same product at a 25 percent return rate.

Amazon reports your advertising attributed sales on orders placed, not on orders that survived. So the denominator of your ACoS is still the full hundred.

Your real break-even ACoS is ₹9,500 divided by ₹60,000.

15.8 percent.

Not 30. And you have been running at 25, congratulating yourself for staying under target.

What that mistake costs, per hundred orders

At a 30 percent ACoS you would spend ₹18,000 to generate ₹60,000 of reported sales. You have ₹9,500 of real contribution to pay it out of.

You are down ₹8,500. That is ₹85 lost on every single order, on a product you believe is at break-even. A thousand orders a month and this is ₹85,000 walking out of the business while the dashboard says you are fine.

This is the specific reason a seller can grow revenue every month and watch the bank balance go the other way. The reporting is not lying to you. It is answering a different question than the one you are asking it.

The formula, so you can run your own

Real break-even ACoS = ( (1 − r) × m − r × c ) ÷ p

Where r is your return rate, m is contribution per order that stays sold, c is what a returned order costs you in logistics and handling, and p is the selling price.

Run our example across return rates and you can see the whole shape of it:

All illustrative, all built on the ₹600 product above. Put your own numbers in.

Two things fall out of that table.

The first is that returns do not nibble at your break-even, they halve it. Going from a clean 5 percent to a fashion-category 30 percent takes you from 27 percent to 13. Every ACoS decision you have ever made sits on the wrong side of that.

The second is the end of the table. On these numbers, at a return rate of about 53 percent, the real break-even ACoS reaches zero. Past that point there is no amount of advertising that is profitable, because the product loses money before a single rupee of ad spend is added. If you have ever had a category where nothing you tried worked, this is worth checking before you blame the campaigns.

What to actually do about it

Work out your real number before you set a target. You need your return rate by ASIN, not a store average. One bad SKU will hide behind nine good ones and quietly fund its own losses.

Treat prepaid share as an advertising lever, because it is one. Most Indian RTO sits on cash on delivery. Prepaid discounts, UPI nudges and removing COD on your worst offenders will move your break-even ACoS further than a week of bid optimisation. It does not feel like an ads job. It has a bigger effect than most ads jobs.

Look at what returns are telling you about the listing. A high return rate on a delivered order is usually the listing having promised something the product did not do. Wrong size chart, colour that photographs differently from how it arrives, a spec buried where nobody read it. That is a content problem showing up as a logistics cost, and it is the same problem that is capping your conversion rate.

Then set the target. Real break-even, minus the net margin you want, is your target ACoS. Not the number from the calculator.

And run it per platform. The arithmetic is a property of your product, but the return rates behind it are not, which is one of several reasons Flipkart is not Amazon with a different logo.

What is a good ACoS?

There isn't one. A good ACoS is any number below your real break-even, and your real break-even is a property of your product and your return rate rather than of Amazon or of your category.

The benchmarks you will find quoted, usually somewhere between 15 and 30 percent, come from American sellers in a market where almost every order stays sold. Applied to an Indian catalogue with a 25 percent return rate they are not conservative, they are optimistic. A 25 percent ACoS is excellent on a product with a 45 percent margin and light returns, and a slow loss on the ₹600 product above.

Work out your own number from the table, then subtract the net margin you want to keep. That is your target. Anything below it is good.

Is a high ACoS good or bad?

Neither, on its own, and the cases where a high ACoS is deliberately correct are the ones worth knowing.

A new product with no ranking and no reviews will run an ugly ACoS while it earns both, and stopping at that point means paying the tuition and dropping out. Defending a branded search term you would rather not hand to a competitor can be worth an ACoS that looks terrible in isolation. A clearance run on ageing stock is trying to convert inventory into cash, not to hit an efficiency target.

What is always bad is a high ACoS nobody chose. If your ACoS drifted up and the explanation is that nobody looked, that is not a strategy, that is a leak with a rationalisation attached.

The honest test is whether you can say what the high ACoS is buying and when it ends. If you can, it is an investment. If you cannot, it is a subsidy.

The honest caveat

Every figure above is illustrative. The ₹160 per return is a reasonable middle for a small parcel and it will be wrong for you. Bulky products cost far more. Some returns come back unsellable, which makes the real number worse than this model, not better.

The point is not the figure. The point is that the standard calculation ignores an input that in Indian e-commerce is large enough to change the answer by half, and almost every seller is optimising against the version without it.

If you want us to run this against your actual margin sheet and your actual return rates, that is the first thing we do on any Amazon engagement, before we set a single bid. And if you are comparing agencies while you are at it, here is what they charge and what each fee model quietly incentivises.

What is a good break-even ACoS for an Indian Amazon seller?

There is no single good number, because break-even is a property of your product and your return rate rather than of Amazon. Calculate it as ((1 − return rate) × contribution per kept order − return rate × cost per returned order) ÷ selling price. For a product with a 30 percent pre-ad margin and a 25 percent return rate, that comes out near 16 percent rather than the 30 percent a standard calculator returns.

Does ACoS include returns?

No. Amazon reports advertising attributed sales on orders placed, so the ACoS denominator still counts orders that were later returned or never delivered. Your reported ACoS therefore looks better than your economics, and the gap widens with your return rate.

How much do returns affect Amazon profitability in India?

Enough to change the decision. Indian marketplace sellers commonly see 20 to 30 percent of orders returned or undelivered depending on category, and each one costs forward and reverse logistics with no sale at the end. On an illustrative ₹600 product that takes real break-even ACoS from 30 percent down to under 16.

How do I reduce RTO on Amazon?

Increase your prepaid share, because most RTO sits on cash on delivery. Discount for prepaid, prompt for UPI, and consider removing cash on delivery on the SKUs and pin codes with the worst history. Then fix the listing gaps that cause delivered orders to come back, which are usually sizing, colour accuracy and a specification the buyer did not see before ordering.