So your ACoS is high. Your team is staring at you. Your CFO has questions. And somewhere, deep inside Seller Central, an auto campaign you set up in 2023 is still spending $4 a day on a search term called "protein powder for dogs."
Welcome to how to lower ACoS on Amazon — the topic every brand owner Googles at least once a quarter, usually around 11pm, usually right after a finance call. [Adjusts imaginary glasses.] Today we're going to actually fix it. Not with a clever tactic. Not with a tool. With the structural moves that quietly drop ACoS 15–30% in any account that's been on autopilot for more than six months.
Quick warning: the answer is not "pause everything that looks expensive." That's the Amazon advertising equivalent of cutting your gym membership to lose weight — technically the spending stops, but you also stop the thing that was actually working.
Let's do this right.
What ACoS Actually Means (and What It Doesn't)
ACoS — Advertising Cost of Sale — is the percentage of ad-attributed revenue your ad spend ate. Spend $100, drive $400 in ad-attributed sales, your ACoS is 25%. Amazon defines it the same way. Simple math, but it carries a giant asterisk.
The asterisk: ACoS only counts revenue Amazon attributes to those specific ads. If your ads built brand awareness and someone bought you organically next week, ACoS does not see that. If your sponsored campaigns lifted your keyword rank and you're now winning page one organically, ACoS does not see that either.
This is why obsessing over ACoS without context is like judging a movie based on the trailer — yes, that's the marketing, but it's not the whole experience. (We'll come back to this. I promise.) For the broader view, you want TACoS — Total Advertising Cost of Sale, which divides ad spend by your total Amazon revenue, organic included. ACoS tells you if individual campaigns are efficient. TACoS tells you if your whole ad investment is building the business.
Now that the definitions are out of the way, let's talk about why your ACoS is climbing.
7 Reasons Your ACoS Is Probably Too High
These are the offenders we see in roughly 80% of brand-owned ad accounts when we open the hood for the first time. None are exotic. All are fixable. Most are the kind of thing that makes you go "oh, right, I knew that."
You're paying for search terms you'd never approve in writing
Auto and broad match campaigns are matching to queries you've never seen. Your negative keyword list hasn't been updated since the Obama administration. Most of the "wasted spend" hiding in your account lives here.
Broad match bids are equal to your exact match bids
Amazon's broad match is famously aggressive. Bidding $2 on a broad keyword is like opening your wallet on a busy street and walking away. Broad bids should sit 20–30% below exact bids on the same theme.
Branded and non-branded keywords are in the same campaign
Brand terms always perform well (your customer was searching for you specifically). Non-brand terms always perform worse (they're discovery). Mixed together, the brand performance masks how badly your non-brand campaigns are doing. Separate them. Always.
Your placement modifiers are at zero
Top of Search converts 2–3× better than other placements in most categories. If your modifier sits at 0%, you're paying the same CPC for every placement and Amazon decides where your money goes. That's not strategy — that's a coin flip.
You set the bids in 2024 and never came back
Conversion rates change. Competitors change. Listing performance changes. A bid set 14 months ago is a bid set for a market that no longer exists. (And yes, I know you've been busy. So has everyone else.)
Daily budgets are flat across all campaigns
Your best campaigns hit their cap at noon and your worst campaigns spend their full allocation on irrelevant clicks until midnight. You're literally capping winners and funding losers equally. Cute.
You're optimizing for ACoS while ignoring TACoS
An efficient campaign that's not building anything is still just an efficient campaign. A "high ACoS" campaign that's driving rank and organic lift is profitable when measured properly. Decide which one each campaign is — then decide what to optimize for.
The Lower-ACoS Playbook (6 Levers)
[Cracks knuckles.] Here's where we stop diagnosing and start operating. Pull these six levers in roughly this order and most accounts will see ACoS drop 15–30% within 60–90 days. The first lever does about half of the work all by itself.
Mine your Search Term Report every 2 weeks
This is the highest-ROI maintenance task in any Amazon ad account. Pull your Search Term Report, sort by spend (descending), and look for two things: terms with high spend and zero conversions (those become negatives), and terms with low spend and high conversions (those get promoted to exact match where you control the bid).
If you do nothing else from this post, do this. Set a recurring calendar reminder for every other Friday and treat it like payroll. Non-negotiable.
Start with the top 20 spending terms in your auto campaigns. Most accounts find at least 5 obvious negatives in the first review. More on this and 9 other common mistakes →
Build your negative keyword list like your margin depends on it
Because it does. Negative keywords are the cheapest, fastest ACoS lever in Amazon advertising — and the most under-used. Every brand we audit has dozens of obvious negatives missing. Read the full guide on building a negative keyword strategy here, but the short version: every account needs negative exact for high-spend zero-converters, negative phrase for irrelevant categories, and negative product targeting for ASINs that consistently take your clicks but never convert.
Add your competitors' brand names as negatives in your branded campaigns. Stops you from accidentally bidding on competitor traffic at exact-match prices.
Separate branded from non-branded keywords — yesterday
If your "Sponsored Products – Manual – Exact" campaign has both your brand name and generic category terms in it, you cannot tell what's working. Brand performance will paper over non-brand inefficiency every time. Separate campaigns. Separate budgets. Separate optimization targets.
Branded campaigns target an ACoS in the 3–10% range (defensive — you're protecting your space). Non-branded targets your true breakeven (around your contribution margin, usually 15–30%). Treat them like different products, because they're different jobs.
Calibrate your placement modifiers (Top of Search especially)
Pull a placement report for any campaign that's been running 60+ days. If Top of Search ROAS is meaningfully higher than other placements (it usually is), raise that modifier. 20–50% is a reasonable starting range. Watch what happens to ACoS over the next 30 days.
Top of Search converts better, but it also costs more. The question is whether the lift in conversion rate more than offsets the higher CPC for your specific category and margin. The placement report tells you. Use it.
Bid by performance, not by feel
The "I'll bid a dollar on everything and see what happens" school of PPC management is, charitably, a school of one. Bids should be set based on your target ACoS, your conversion rate for that keyword, and the historical CPC range. The formula isn't complicated: max bid = (sale price × target ACoS) ÷ (assumed CPC × conversion rate). Run the math for your top 20 keywords. Update bids weekly until they stop drifting.
If you're managing meaningful spend ($10K+/month), tools like Pacvue, Perpetua, or Adtomic can automate this. They are not magic. They are a calculator with a UI. But a calculator with a UI is still useful when you have 400 keywords to manage.
Tighten your campaign structure (one campaign, one ad group, one theme)
Stuffing 100 keywords into one ad group means Amazon's budget algorithm decides which keywords "win" — not you. A tighter structure (one campaign per match type per theme) creates more campaigns to manage but dramatically better control over individual bids. We break this down in the full PPC optimization guide.
The tradeoff: more campaigns = more cognitive load. The fix: a strict naming convention. Brand | Product | Match Type | Theme. Use it religiously and 40 campaigns feels like 4. Skip it and 4 campaigns feels like 40.
When High ACoS Is Actually Fine
OK — pause before we start swinging the ACoS-lowering hammer at everything. Sometimes a "high ACoS" campaign is doing the most important work in your whole account.
A campaign launching a new product won't hit your target ACoS for weeks. It's supposed to be unprofitable on a per-transaction basis during launch — you're buying ranking and reviews. The 2026 Amazon playbook covers this in more detail, but the short version: ACoS at 60% during launch is not a campaign problem. It's an investment.
Same goes for category-conquest campaigns targeting your strongest competitors' branded terms. ACoS will look terrible. But if you steal even a small percentage of buyers who would've gone to a competitor, that customer is yours forever (and their LTV is enormous). Don't kill the campaign because the snapshot is ugly.
Before pausing any high-ACoS campaign, ask three questions: Is it building organic rank? Is it driving discovery for a new product? Is it defending share against a competitor? If the answer to any of those is yes, the ACoS number isn't telling you the whole truth. Check TACoS instead.
The Quarterly ACoS Audit Checklist
Even well-managed accounts drift. Campaigns get launched, old ones get forgotten, naming conventions slowly become a war crime. A quarterly audit is the reset that keeps everything intentional — and it's where most of our long-term ACoS reductions actually happen.
When to Call for Backup
Look. Most brands can run this playbook themselves if they put 4–6 hours a week into it. We've written it down precisely so you can. Our free PPC analyzer will even surface the obvious negatives for you in under 60 seconds.
But — there's always a but — if your spend is past $20K/month, you're running multi-brand, your team is spread across other channels, or you've simply run out of bandwidth to do this consistently, that's when a fractional senior team starts to make sense. We break down the math on hiring vs. fractional vs. agency here. Short version: $240–300K/year for senior judgment running every lever in this post, every week, on your account. About half what hiring the same team internally would cost.
Or just keep running the playbook yourself. Either works. The point is to actually run it — not to add it to the list of things you'll definitely get to next month. (You won't. Nobody does. That's why ACoS is high in the first place.)
If you take one thing from this post: the search term report is the single most powerful tool in your account, and almost nobody uses it consistently. Pull it every other Friday. Mine the negatives. Promote the winners. The other five levers will multiply that work — but the first one alone does about half the job.
[Takes a bow.] That's the playbook. Now go open Seller Central. (After you finish this coffee. We're not animals.)
FAQ
A good ACoS depends on your contribution margin. If your product margin is 35%, anything under 35% ACoS is breakeven or profitable on a transaction basis. The real benchmark isn't a number — it's whether your blended TACoS is trending down while revenue grows. Most categories operate in the 15–30% ACoS range for healthy long-term campaigns.
The five most common causes: wasted spend on irrelevant search terms (no negative keywords), bidding too aggressively on broad match, mixing branded and non-branded keywords in the same campaign, leaving placement modifiers at zero, and never auditing campaign budgets. Most accounts can drop ACoS 15–30% just by fixing the first two.
ACoS = ad spend ÷ ad-attributed revenue, expressed as a percentage. If you spent $100 in ads and they drove $400 in attributed sales, your ACoS is 25%. ACoS only counts revenue Amazon attributes to those specific ads — it ignores organic sales the ads may have lifted. That's why TACoS is the more honest metric.
ACoS measures ad spend against ad-attributed revenue only. TACoS measures ad spend against your total Amazon revenue, including organic. ACoS tells you if individual campaigns are efficient. TACoS tells you if your overall ad investment is building the business or just buying sales. A healthy account has TACoS trending down over time while revenue grows.
Quick wins (negative keywords, budget reallocation) can show up in 2–4 weeks. Structural changes (rebuilding campaign architecture, separating branded from non-branded) take 6–12 weeks to settle. Anything claiming a permanent ACoS drop in under 30 days is either pausing winning campaigns or borrowing performance from organic — neither is sustainable.
Not without context. A "high ACoS" campaign that's building ranking and driving organic lift is profitable when you measure TACoS. A "high ACoS" campaign with no organic effect is just expensive. Look at the trend, the search terms it's converting on, and whether the keywords are building rank — then decide. Pausing is a hammer; you need a scalpel.