How to Lower ACoS Without Killing Sales

Practical bid and search term tactics that improve ad efficiency without shutting down growth.

The fastest way to lower your ACoS is also the fastest way to destroy your rank: cut all your bids by 30% across the board. Your ACoS drops. So does your impressions count. So do your clicks. So do your sales. Within two weeks, Amazon interprets your falling sales velocity as a signal to demote your organic rank. You've solved the ACoS problem by creating a revenue and ranking problem. What you actually need is a surgical approach — find the waste, eliminate it precisely, and leave everything profitable untouched.

Understand What's Driving Your ACoS

Before you touch a single bid, you need a diagnosis. ACoS (Advertising Cost of Sale) is calculated as:

ACoS = Ad Spend ÷ Ad Revenue × 100

If you spent $200 on ads and generated $800 in ad-attributed revenue, your ACoS is 25%. Whether that's good or bad depends on your margins — use the FBA Profit Calculator to find your break-even ACoS before deciding what threshold you're actually optimizing toward.

High ACoS has three root causes, and which one is your problem determines everything about how you fix it.

Root cause 1: Wrong keywords. Your campaigns are serving ads against search terms that don't match your product or buyer intent. Shoppers click but don't buy because they were looking for something different. Every click costs you money and generates no revenue.

Root cause 2: Poor listing conversion rate (CVR). Your keywords are relevant, the clicks are from real buyers, but your listing isn't closing the sale. Your ACoS looks high because you need a lot of clicks per sale — and more clicks per sale means higher spend per unit sold.

Root cause 3: Mismatched bids. You're overpaying per click on terms where your conversion rate doesn't justify the bid. You might be paying $2.00 per click on a term that converts at 4% on a $25 product — the math doesn't work.

How to diagnose your specific problem:

Open your Search Term Report (Seller Central → Reports → Advertising Reports → Search Term Report). Sort by spend descending. Look at the CVR column for your top 10 spending terms.

• If the high-spend terms have CVR under 5% and are clearly irrelevant queries — Root Cause 1 (wrong keywords).
• If the high-spend terms are relevant and have reasonable CTR but CVR is under 5% for all of them — Root Cause 2 (listing problem).
• If the terms are relevant, CVR is 8–12%, but CPC is very high relative to your selling price — Root Cause 3 (bid mismatch).

Most sellers with high ACoS are dealing with Root Cause 1 as the primary driver. Auto and broad campaigns are running discovery queries — "what is," "how to," color/size variants you don't sell, competitor brand names — and those terms are clicking and spending without ever converting.

Layer 1: Search Term Isolation

This is the single highest-leverage tactic for lowering ACoS without touching your winners. The idea is simple: find the terms in your auto and broad campaigns that are actually converting, then move them into exact match campaigns where you control the bid precisely. Stop letting Amazon's match algorithm overspend on loose variants of your best terms.

Step-by-step:

1. Download your Search Term Report for the last 30–60 days.
2. Filter to rows where Clicks ≥ 10 AND Orders ≥ 1. These are terms with enough data to be statistically meaningful and at least one proven conversion.
3. For each qualifying term, note the CPC (cost per click) from the auto or broad campaign where it's currently spending.
4. Add each of these terms as an exact match keyword to a dedicated exact match campaign. Set the starting bid at CPC from auto × 0.9. You're starting slightly below what you were paying in the uncontrolled environment — you'll adjust from there.
5. Add these same terms as negative exact to your auto campaign. This forces all future clicks on those terms to route through your exact campaign, where you control the bid.

Why this works:

In an auto campaign, Amazon's algorithm decides in real time how much to bid for each impression — and it often overbids when it thinks a click is likely, even if that click won't convert for your specific listing. When you move a proven term into exact match with a fixed bid, you're paying what you've decided the term is worth rather than what Amazon's algorithm decided. Over 30 days, this alone can reduce ACoS by 8–15 percentage points on your best campaigns.

The harvesting loop: Run this process every two weeks. Every two weeks, you're pulling more proven terms out of your discovery campaigns and into controlled exact match campaigns. Over 60–90 days, your exact campaigns grow into efficient, scalable revenue drivers while your auto campaign shrinks proportionally. This is the core mechanic of a mature PPC account. For the step-by-step cadence, see the Amazon PPC keyword harvesting routine.

Layer 2: Negative Keyword Waste Removal

Once you've harvested your winners, turn your attention to the active waste. These are the search terms eating your budget every day with zero contribution to revenue.

Types of wasted spend to look for:

• Irrelevant product type queries — You sell a stainless steel water bottle. Amazon's auto campaign is serving your ad for "plastic water bottle" and "glass water bottle." Different shoppers, different intent, and your product doesn't match what they're looking for.
• Branded competitor terms — If you're bidding on competitor brand names and not converting, you're donating to Amazon. Unless you're specifically running a conquest strategy with a proven conversion rate, these should be negated.
• Research queries — "What is the best water bottle" or "water bottle comparison" indicates a shopper early in the research phase. These rarely convert well for direct product ads.
• Size/color variants you don't sell — If you only sell a 32oz bottle, "16oz water bottle" and "64oz water bottle" are waste. Every click from those queries is from a shopper who needs something you don't have.

How to find them:

In your Search Term Report, sort by spend descending. Work down the list. Any term with more than $5 in spend and zero conversions over 30 days is a candidate for negation — unless it's a brand new term that hasn't had time to generate data.

Where to add negatives:

• Auto campaign: add irrelevant specific queries as negative exact. You want to block that precise query without accidentally blocking related relevant queries.
• Broad match campaigns: add pattern-based waste as negative phrase. For example, if your product doesn't come in red, add "red" as a negative phrase — this blocks "red water bottle," "water bottle red," etc. without needing to add each variation individually.

Maintenance cadence: A bi-weekly negative keyword audit takes 15 minutes once you know what to look for. Pull the STR, sort by spend, scan the top 20–30 terms. You're not doing a complete overhaul — you're trimming the consistent wasters. Sellers who do this every two weeks consistently report 20–30% reductions in wasted spend within 60 days.

Layer 3: Bid Adjustments Without Killing Volume

Once you've isolated your winners into exact campaigns and eliminated identifiable waste with negatives, you're ready to adjust bids. Not before. Cutting bids before completing Layers 1 and 2 is how sellers accidentally reduce spend on their profitable terms while leaving the waste intact.

The formula for the right bid:

Max CPC = Target ACoS × Selling Price × Conversion Rate

Example: Your target ACoS is 25%. Your selling price is $30. Your conversion rate on a given term is 8%.

Max CPC = 0.25 × $30 × 0.08 = $0.60

If you're currently paying $1.10 per click on that term, you're structurally overpaying. You'd need a CVR of ~14.7% at $1.10 CPC to hit 25% ACoS on a $30 product. If the real CVR is 8%, the bid needs to come down to $0.60.

Reduce bids on specific terms — not at the campaign level. Go into your exact match campaigns, find the individual keywords with ACoS above your target, and reduce those specific bids. Do not touch the keywords hitting your target or below. A blanket campaign-level bid reduction will cut your profitable terms alongside the underperformers.

Placement bid adjustments. Check your campaign placement data (Campaign Manager → select campaign → Placement tab). If "Top of Search" placement is generating 70% of your spend but only 40% of your sales, the top placement modifier is likely inflating spend beyond what that placement deserves. Reduce the top-of-search modifier by 10–20% and monitor for 14 days. Conversely, if "Product Pages" placement is your most efficient placement but it's only getting 15% of your spend, consider a positive modifier there.

Dynamic bids — switch to "Down only" on high-ACoS campaigns. If a campaign's ACoS is above your target and it's set to "Up and Down" dynamic bidding, Amazon is raising your bids in real time to win impressions it thinks will convert. Given that the campaign is already converting poorly, Amazon's auction-time bidding is making a bad situation worse. Switch to "Down only" — Amazon can reduce your bids when conversion looks unlikely, but it can't raise them above what you've set.

This is also the point where starting with the right keyword strategy pays dividends you can't retroactively buy. LaunchFast builds your keyword targeting before launch, which means your campaigns open on purchase-intent terms with realistic CVR expectations rather than on a discovery mess where you're trying to reverse-engineer a reasonable ACoS from a structurally broken starting point.

Layer 4: Listing CVR Fix

Here's the check that most PPC guides skip: if your conversion rate is genuinely low — under 5% — no amount of bid optimization will fix your ACoS. The math won't work regardless of CPC because you need too many clicks per sale.

CVR benchmarks to know:

If your CVR is below 5% and your keywords are relevant, the listing is the problem. Cutting bids will just reduce impressions and lower your organic rank — it won't improve the underlying conversion problem.

Quick listing CVR diagnostic:

• Main image: Does it stand out in search results against the products next to it? Pull up your main keyword in Amazon's search bar and look at the results. Is your image cleaner, more compelling, and more product-focused than the competition? If not, this is likely your CVR killer.
• Price: Are you within 10–15% of the page-one average for your keyword? Significant price premiums without obvious quality differentiation will tank CVR.
• Review count and rating: Under 15 reviews, many shoppers won't buy regardless of everything else. Under 4.2 stars, conversion drops sharply. This is a timing and product quality issue, not a PPC issue — but it affects your PPC math directly.
• Bullet points: Are they answering the questions a buyer actually has — dimensions, compatibility, materials, use cases — or are they generic marketing copy? Shoppers who click your ad and can't quickly find the information they need to make a purchase decision will bounce.

Do not reduce bids to compensate for poor CVR. You'll end up with fewer impressions, lower sales velocity, lower organic rank, and the same underlying listing problem. Fix the listing first, then optimize bids. The Amazon Listing Optimization Basics guide covers the specific elements — images, bullets, and title — that move CVR most.

For context on how CVR affects your overall profitability, not just ACoS, run your numbers through the FBA Profit Calculator.

What NOT To Do

Pausing entire campaigns to "save money." When you pause a Sponsored Products campaign, Amazon's algorithm loses the sales velocity signal that campaign was generating. If the campaign was contributing even a handful of daily sales, organic rank can start declining within 3–5 days. The right move is almost never to pause — it's to isolate the specific wasteful terms within the campaign.

Cutting all bids by a flat percentage at once. If you reduce all bids by 20%, you're cutting your winners, your mediocre performers, and your waste at the same rate. Your profitable terms lose impression share. Your waste gets mildly less wasteful but keeps spending. The net effect is usually lower sales volume with a marginal ACoS improvement that wasn't worth the tradeoff.

Optimizing ACoS without watching TACoS. TACoS (Total Advertising Cost of Sale) divides your ad spend by your total revenue — both ad-attributed and organic. As your PPC drives rank improvements and organic sales grow, TACoS should decline even if ACoS stays flat. A seller with 25% ACoS but 8% TACoS is in a healthy position. Watch both numbers, not just ACoS in isolation.

Making changes daily. PPC data is noisy over short windows. A term that looked like a waste on Day 3 might be your best converter on Day 10. Each bid change or negative keyword addition needs at least 7–14 days of post-change data before you can evaluate the impact. If you're adjusting daily, you're making decisions on noise rather than signal.

If you're new to setting up the underlying campaign structure before you start optimizing, the Amazon PPC for Beginners: First Campaign Setup guide covers the exact three-campaign foundation that makes these optimizations possible.

60-Day ACoS Improvement Plan

This timeline assumes you're working with a campaign account that's been running for at least two weeks before you start. If you're optimizing a brand-new account, shift everything right by two weeks to allow for initial data collection.

For category-specific ACoS benchmarks — what's actually "good" for your niche — see the ACoS benchmark targets guide. Those benchmarks matter because optimizing to 20% ACoS in a category where 35% is average and profitable is overtightening — you're leaving revenue on the table.

Done in sequence, these four layers address every root cause of high ACoS without the blunt-force approach that tanks your rank. Isolate your winners, eliminate your waste, price your bids mathematically, and fix the listing if the listing is broken. In 60 days, you'll have an account that converts efficiently and a clear picture of what to scale.