Lowering ACoS is the most requested thing we hear from brand owners, and the most misunderstood. Anyone can drop ACoS by slashing bids. The problem is that the brands who do this usually watch their sales fall right alongside it, and end up worse off than when they started.
The goal was never a lower ACoS in isolation. The goal is more profit. Those are not the same thing, and confusing them is the single most expensive mistake we see in Amazon advertising.
Start with wasted spend, not all spend
Open your search term report and you will almost always find the same pattern: a small number of search terms quietly burning budget with clicks and no orders. This is the first place to cut, because it lowers ACoS without touching a single dollar of revenue.
We sort every search term into four buckets:
- Harvest — converting terms worth promoting into their own targeted campaigns.
- Negate — clicks with spend and no orders, added as negative keywords.
- Promote — strong performers that deserve a higher bid.
- Watch — not enough data yet, leave them alone.
Negating the waste alone often drops ACoS by several points in the first month, with zero impact on sales.
Protect your profit drivers
Every account has a handful of keywords that do the heavy lifting. These are the terms where you rank, convert, and win. The instinct when cutting ACoS is to trim everything equally. Do not. Cutting bids on your best converters to save a few points of ACoS is how you quietly hand sales to a competitor.
Cutting ad spend is easy. Cutting wasted spend while protecting your best-performing keywords is the real skill.
Set a target ACoS that reflects your margin
Your target ACoS is not a vibe, it is a number derived from your contribution margin. A product with healthy margins can profitably sustain a higher ACoS than a thin-margin product. When the target is set correctly, "lowering ACoS" stops being a guessing game and becomes a precise lever.
The simple math
If a click is worth a certain amount in revenue, and you know your target ACoS, your bid almost sets itself: revenue per click multiplied by target ACoS. Build bids from that number with sensible guardrails and you stop overbidding on weak terms and underbidding on strong ones.
Give changes time to settle
Amazon's algorithm needs data. Brands that change bids every day never let the system stabilize, then mistake noise for signal. Make deliberate changes, then give them one to two weeks before judging the result.
Done right, lowering ACoS is not about spending less. It is about spending smarter, so every dollar works harder and the profit, not just the percentage, actually improves.