PPC

Most PPC waste is a math problem, not a bidding problem

Before touching bids, work out each SKU's break-even ACoS from real fees and landed cost. Then the wasted spend in your search term report becomes obvious.

September 25, 2026 · 7 min read

Sponsored Products waste is usually diagnosed as a bidding problem: bids too high, match types too loose, not enough negatives. Those are real, but they are downstream. The more common root cause is that the seller never wrote down how much advertising each SKU can actually afford, so there is no line to judge any keyword against.

If you do not know a SKU's break-even ACoS, a 35% ACoS campaign could be quietly profitable or losing money on every order, and the dashboard looks identical either way. This post walks through fixing the math first, then using it to find the spend that is genuinely wasted.

Step one: break-even ACoS per SKU, not per account

Break-even ACoS is the share of the selling price you can spend on ads and still make zero profit on an ad-attributed sale. It is simply your pre-ad profit margin:

Break-even ACoS = (price − referral fee − FBA fulfillment fee − landed cost of goods − other per-unit costs) ÷ price.

Two things make this harder than it looks. First, the fees must be the fees you are actually being charged, not the ones you estimated at launch — size tier reassignments and fee schedule changes move this number without warning. Second, landed cost has to include freight, duties, prep, and inbound placement costs, not just the factory invoice. Leave those out and every break-even number is inflated, which means every campaign looks healthier than it is.

Example: a $30 item with a $4.50 referral fee, $6.20 fulfillment fee, and $9.00 landed cost leaves $10.30 before ads — a break-even ACoS of about 34%.
The same product at $24 after a price cut leaves $5.20 — break-even drops to about 22%. The campaign did not change; the ceiling did.
Do this per SKU and per variation. A size or color that lands in a heavier fee tier can have a completely different ceiling from its siblings.

Step two: set a target below break-even, on purpose

Break-even is the line where ads stop destroying profit, not a goal. Decide per SKU what the ads are for. A mature SKU that should throw off cash needs a target well under break-even. A launch SKU where you are buying ranking and reviews can justify running at or above break-even for a defined period — but that should be a written decision with an end date, not a drift nobody noticed.

Writing the target next to the break-even number is the single most useful thing you can do. It turns every future optimization session from “does this feel expensive?” into a comparison against a number.

Step three: find the waste in the search term report

With per-SKU ceilings in hand, pull the search term report for a meaningful window — long enough that each term has had a fair chance to convert. Then sort terms into buckets:

Spend with zero orders past a threshold — pick a threshold tied to your economics, such as spend equal to one or two orders' worth of pre-ad profit. Terms past it with no sale are the cleanest negatives you will ever find.
Orders, but ACoS above break-even — these convert but at a loss. Lower bids or isolate them into their own campaign with a tighter budget rather than negating outright.
Irrelevant intent — terms for a different product type, a competitor's exact brand, or a use case you do not serve. Negate these even if they occasionally convert; they are usually returns in disguise.
Converting under target — harvest these into exact-match campaigns so you control the bid instead of letting an auto or broad campaign set it.

The wastes nobody looks at

Beyond obviously bad search terms, a few patterns burn money quietly:

Advertising SKUs that cannot sell. Ads running on a listing that is out of stock, stranded, or suppressed spend nothing — but ads running on a listing about to stock out push the last units at an ad cost, then lose the ranking you paid for anyway. Pause or throttle ads as inventory runs down.
Advertising a variation that lost the Buy Box. If another seller or a pricing issue has taken the Buy Box, many ad placements simply stop serving or serve poorly. Fix the Buy Box before diagnosing the campaign.
Duplicate targeting. The same keyword in an auto campaign, a broad campaign, and an exact campaign means you are bidding against yourself and splitting the data three ways. Negate exact-match winners out of the discovery campaigns.
Bidding on your own brand with no competitor pressure. Brand defense can be worth it, but if nobody else is bidding on your brand terms, you may be paying for sales you would have received organically. Test pausing it on one SKU and watch total orders, not ad orders.
Stale ceilings. A price change, a fee change, or a new supplier invoice changes break-even overnight. Campaigns tuned to last quarter's margin are tuned to the wrong number.

Judge ads on total profit, not ACoS alone

ACoS only looks at ad-attributed sales. Ads also lift organic ranking, which is why some sellers watch TACoS — ad spend divided by total sales for the SKU. Neither ratio is profit. The real question each month is simple: after fees, cost of goods, and ad spend, did this SKU make more money than it did last month, and is the trend going the right way?

If a SKU's ad spend rises while its total profit falls, the ads are not working no matter what the ACoS says. If ACoS rises but total profit rises with it because organic sales grew, that may be fine. You can only tell with profit per SKU in front of you.

A monthly routine that fits in an hour

Refresh break-even ACoS for your top SKUs using current fees and current landed cost.
Flag any SKU whose break-even moved more than a few points since last month and revisit its targets.
Pull the search term report, negate the zero-order terms past threshold, and harvest the winners into exact match.
Check inventory and Buy Box status for every advertised SKU before raising any budget.
Write one line per SKU: target, actual, what you changed. Next month you compare against the note, not your memory.

Where MySellerHQ fits

MySellerHQ is built around the margin side of this equation. It reads orders, fees, and inventory from Amazon's Selling Partner API with read-only permissions and shows profit per SKU using the fees you were actually charged, so a fee or size-tier change that moves your break-even point shows up instead of hiding in a settlement report. It also surfaces stock and Buy Box issues that make ad spend pointless.

It does not manage your campaigns, change bids, or edit listings — writes are off by design. The point is that when you sit down to optimize ads, the ceiling for each SKU is already correct.