Nobody tells you this when you start running Amazon Ads, but the first time ACoS spikes and you can’t explain why, you don’t fix it, you “guess” at it.
You try lowering a few bids. You then try to pause a keyword that feels suspicious. You check if anything was paused by accident. Nothing happens. So you try a different explanation.
Maybe it’s seasonal, maybe a competitor is doing something, maybe the algorithm changed. You make another adjustment and wait for a few days. But the ACoS doesn’t stop climbing.
The problem is that Amazon gives you the outcome number — ACoS — but not the chain of events that produced it. And when you don’t have the chain, you make changes based on intuition rather than a proper analysis.
Sometimes, Amazon ACoS spikes get “fixed” that way, while some recover on their own anyway. But there’ll be some that keep getting worse.
This is the diagnostic framework that traces backward from ACoS to the cause, so you’re not adjusting things at random.
Why is my Amazon ACoS suddenly high? (And why is it so hard to figure out?)
ACoS is Advertising Cost of Sale.
It’s calculated as:
ACoS = Ad Spend/Ad Revenue
And it’s expressed as a percentage. So a 30% ACoS means you spent $30 to generate $100 in attributed sales. The higher the ACoS, the less efficient your advertising, which means you’re spending more for each dollar of revenue you get back.
Here’s where things get complicated: ACoS can rise for reasons that are completely unrelated to each other.
That’s the core diagnostic problem. Amazon’s native reporting shows you that ACoS moved. It does not show you which side of the equation drove the movement or what caused it. And the platform’s column-based reporting is built for campaign management, not causal analysis, so it surfaces metrics rather than the relationships between them.
There’s also a timing problem layered on top. As Optmyzr’s research across 14,991 Amazon Ads campaigns found, attributed sales figures in the top 5% of campaigns grew by at least 18.75% between Day 1 and Day 17 of the same reporting period; meaning any data from the last two to four weeks should be treated as provisional. If you’re diagnosing an ACoS problem using recent data, you may be reacting to a number that’s still moving.
Scott Desgrosseilliers, CEO of Wicked Reports, addressed this directly in a PPC Town Hall episode:
“I find that time lag is real and not a lot of people understand it. And it’s a big advantage — it’s a competitive advantage if you do, ‘cause you can spend where audiences look like they’re not converting and, you know, ignore some other ones that are inflated.”
This is the implication for Amazon’s diagnosis. Before acting on an ACoS spike, confirm that the data window you’re looking at is fully settled.
What are the three root causes of an Amazon ACoS spike?
Every ACoS spike traces to one of three structural causes. Getting clear on which one you’re dealing with is what determines whether the fix actually works.
Cause 1: Is your ACoS high because of a traffic problem?
ACoS is spend divided by revenue. If spend is rising without a matching rise in revenue, start with what’s happening to the traffic.
If impressions dropped, your ads are winning fewer auctions.
The most common reasons:
- A bid that’s no longer competitive because a competitor entered the category or increased their spend
- A daily budget that’s hitting its cap early and cutting your visibility for the rest of the day
- A keyword relevance shift that’s reducing how often Amazon considers your ad eligible
The fix here is bid adjustment, budget reallocation, or expanding your keyword set.
If impressions held but click-through rate fell, your ads are showing but shoppers aren’t clicking.
This usually means one of three things:
- Your ad creative or main listing image has become less competitive relative to what’s showing up alongside it;
- The search terms triggering your ads have shifted toward lower-intent queries; or
- Your placement mix has tilted, which results in more impressions landing on product pages where CTR is structurally lower than top-of-search.
Before adjusting bids, identify which of these is the actual driver.
If CTR held but cost-per-click rose, you’re dealing with auction pressure. A competitor entered your keyword set, or existing competitors increased their bids.
If CPC went up without a corresponding improvement in conversion rate, ACoS rises automatically and becomes a margin compression from competition.
Cause 2: Is your ACoS high because your conversion rate dropped?
This is the most misdiagnosed cause of ACoS spikes, and the one where wrong interventions do the most damage.
When conversion rate drops, the reflex is to adjust bids, tighten targeting, pause low-converting keywords, and restructure campaigns. These actions reduce traffic volume. They do not fix the underlying problem if it is on the product side.
A conversion drop that manifests as an ACoS spike can trace back to:
- Reviews dropping below a threshold that shoppers trust;
- A competitor launching a better product or undercutting your price;
- Listing images or copy becoming less competitive over time;
- Stock running low enough that delivery windows extended and shoppers chose a faster alternative; or
- The buy box being lost to another seller.
None of these are problems that bid changes will solve. They require listing-level intervention first.
The placement mix is a related version of the same problem. Top-of-search converts at a meaningfully higher rate than product pages or rest-of-search. If budget allocation shifted toward lower-quality placements over time — without any intentional decision you made — blended conversion rate will fall and ACoS will rise even if nothing in your targeting structure changed. Pulling the placement report is a necessary step before concluding anything about campaign performance.
Here’s another point to remember: optimizing purely for ACoS while conversion rate is quietly eroding. ACoS is an ad efficiency metric.
TACoS — total advertising cost of sale, which measures ad spend against total revenue including organic — is a much more honest read on whether your business is actually healthy.
You can have a tightening ACoS and a deteriorating TACoS at the same time, and the second number is the one that tells you whether your organic foundation is holding up. If you’re only watching ACoS, you can be making everything look fine in the ads console while the real business is slowly getting worse.
Cause 3: Is your ACoS high because of a demand or market shift?
This is the cause people most often try to fix with PPC changes, and the one where they have the least effect.
- Seasonality
Seasonality is the most straightforward version: demand for your category dropped, and no bid optimization will generate purchase intent that isn’t there.
- Competitor landscape
If a major competitor in your niche ran a big promotion or slashed their price, your CVR will fall across the board. It’s not your listing or your keywords, but shoppers are just doing the math differently now.
The reverse is also true: if that competitor runs out of stock and pulls their campaigns, you’ll suddenly have more impressions available and start spending faster than you expected. These are market dynamics, not campaign problems, and they require reading the competitive landscape rather than pulling another report.
- Organic rank degradation
When organic rank falls — because of a stockout, a sudden review drop, or a pricing shift that reduced sales velocity — PPC ends up doing more work to maintain the same revenue level.
Spend stays the same or rise while organic-assisted revenue falls away, and ACoS deteriorates even if every campaign decision was correct. Adjusting bids doesn’t fix this; the only way is to recover the organic rank.
- Branded search volume
ACoS is highly sensitive to the branded/non-branded traffic mix in your account. If branded search volume declined — because a competitor’s brand gained traction, because an external event affected brand perception, or because seasonal interest simply dropped — your blended ACoS will rise even if non-branded campaigns are performing exactly as expected.
Treating a brand health problem as a campaign performance problem leads to changes that don’t move the right number.
How do you diagnose an Amazon ACoS spike? (A 6-step investigation)
The framework above identifies the causes. This sequence is how you determine which one you’re dealing with, using data that’s actually reliable.
Step 1: Is your comparison window actually clean?
This is the step most teams skip, and it’s why a lot of ACoS diagnoses end up wrong. Amazon’s attribution window spans 7 days for Sponsored Products and 14 days for Sponsored Brands and Sponsored Display. Data in that window is still moving, conversions are being attributed, returns are being processed, traffic is being validated.
If your comparison period sits inside that window, you are comparing settled data to provisional data, and the delta you’re seeing may not be real.
Choose a comparison window that ends at least 14–28 days before today. Accept the slight lag in exchange for working with numbers that won’t change on you by next week.
Step 2: Did spend rise, or did revenue fall — or both?
Before looking at any other metric, confirm which side of ACoS moved. Spend up with flat revenue is a Cause 1 or Cause 3 signal. You’re generating traffic, but it’s either more expensive or it’s meeting weaker demand.
Revenue down with flat spend is a Cause 2 or Cause 3 signal. Traffic is arriving but not converting. Both sides moving adversely at once is the worst version and often involves multiple causes.
Step 3: Check impressions and CTR by campaign type, not blended
Sponsored Products and Sponsored Brands have different placement types and attribution windows. Do not blend their performance when diagnosing.
If impressions dropped, start with bids, daily budget caps, and keyword relevance. If impressions held but CTR fell, look at placement mix and ad creative.
You can run this check independently for each campaign type before drawing conclusions.
Step 4: Is CVR the problem?
If CTR held but ACoS is rising, conversion rate (CVR) is almost certainly the issue.
Pull CVR by ASIN or product group, not just by campaign. If the CVR drop is concentrated in specific ASINs, go to those listing pages before touching a single bid. Check reviews, buy box ownership, price relative to current top-of-search results, and whether low stock is extending delivery estimates.
Step 5: Have you pulled the placement report?
Break performance by placement: top-of-search, product pages, rest-of-search. If CPC rose on top-of-search without a CVR improvement, that’s auction pressure — a Cause 1 problem with a bid response.
If your share of product page impressions grew, blended CVR will typically fall even without any campaign changes you made. This report is the one most teams skip when diagnosing ACoS spikes, and it is frequently where the answer is hiding.
Step 6: Is the ACoS spike worse in branded or non-branded campaigns?
Pull branded and non-branded campaigns and compare their ACoS movements independently.
If the deterioration is concentrated in branded campaigns, the diagnosis shifts toward brand health and organic rank (Cause 3). If it is concentrated in non-branded campaigns, the problem is more likely in Causes 1 or 2.
Treating these as a single blended pool produces misleading averages that hide which problem you’re actually solving.
Why most teams skip to the wrong step
Running this six-step investigation manually means pulling at minimum four separate Amazon reports — campaign performance, search term report, placement report, advertised product report — aligning date ranges that may not match, cross-referencing metrics in spreadsheets, and building the causal chain by hand.
Most teams under time pressure skip steps 3 through 5. They start at “ACoS is high” and go straight to “adjust bids.” That’s why the same accounts have the same ACoS problems month after month.
Dustin Miller, founder of The PPC Pros, described the manual version of this investigation plainly: “If you’ve ever tried to find out where a problem is happening… that’s called a time suck.”
His team used to spend 30–60 minutes per investigation, sometimes longer on enterprise accounts. Over the years using Optmyzr, The PPC Pros have saved 5,000+ hours, and as Dustin noted, much of that came from the dashboard and PPC Investigator alone, before the team even got into automation.
How does Optmyzr help you diagnose and fix Amazon ACoS problems?
Running this diagnostic manually takes time most teams don’t have. That’s where Optmyzr closes the gap. Here’s what each tool does and where it fits in the framework above.
PPC Investigator: Trace the ACoS spike backward through the metric chain
Optmyzr’s PPC Investigator gives you a Cause Chart that maps the relationships between metrics in your account. So instead of staring at a single ACoS number and guessing, you can see where in the performance chain the movement started.
The way it works: you pick the metric you’re investigating and a date range, and the Cause Chart visualizes potential causality across the metric hierarchy.
You can compare clicks, ad sales, spend, CTR, CPC, and ACoS to track back its rise or fall. For this example, let’s analyze how ACoS can be tracked. If ACoS spikes, it could be because of spend or ad sales, and let’s assume it’s because of a drop in spend, which could mean a drop in CPC or a rise in clicks, which could then be drilled down to CTR and impressions.
This Cause Chart analysis helps you figure out exactly what went wrong or right in just a few minutes. You can combine this with the Performance Comparison tool and Magic Quadrant to get a holistic view of your campaign performance and ACoS spike.
The Performance Comparison tool complements the Cause Chart by letting you compare any two custom date ranges side by side across campaigns and ad groups, with the percentage change calculated automatically. This is how you confirm that the window you’re comparing is apples-to-apples before committing to a diagnosis.
When the ACoS problem is spread across a whole portfolio, Magic Quadrant plots your top campaigns, keywords, or ad groups across two axes you define. You immediately see what to scale, what needs fixing, and what to pause, instead of sorting through 100 rows in a spreadsheet to find the campaigns dragging the average down.
When you have a portfolio-level ACoS problem, this is how you identify which specific campaigns are pulling the average down without having to sort through a hundred rows in a table.
Sidekick: Ask your ACoS question in plain English and get a contextual answer
While PPC Investigator traces the structural chain, Optmyzr’s Sidekick handles the interpretation and narrative layer, the part where you need to understand what a pattern means and what to do next.
Sidekick lives inside Optmyzr, not in a separate tab, not in a generic AI tool that has no access to your account.
So when you open the Sidekick panel on your Account Dashboard and type “Why did ACoS go up this month?”, it reads the data in the view you’re currently in, accesses your account’s campaign data, keywords, and ads, and gives you a contextual answer specific to your account.
KPI Alerts: Know before the ACoS spike becomes a problem
The diagnosis framework above is for when ACoS has already spiked and you need to understand why. But the better version of this problem is catching the movement early enough that you’re investigating a 3-point rise, not a 15-point collapse.
You can set threshold-based alerts on ACoS, spend, conversions, CTR, CPC, impressions, ROAS, and other supported metrics at the account or campaign level, delivered via email, Slack, or Microsoft Teams.
As Frederick Vallaeys, Optmyzr’s CEO, wrote in the data delay study:
“If you’re managing budgets or evaluating ROAS based on a snapshot that’s still shifting, you could be pausing profitable campaigns or under-crediting what’s actually working.”
Measuring the full metric chain, not just the outcome number, is what separates teams that manage ACoS proactively from teams that explain it after the fact.Alerts, Cause Charts, Sidekick, Performance Comparison, and Magic Quadrant are the tools that make measurement operationally manageable rather than theoretically desirable.
Trying to diagnose ACoS spikes manually across multiple reports every month? Optmyzr’s PPC Investigator, Sidekick, and KPI Alerts can give you the diagnostic chain in one place.
Frequently Asked Questions
What is a good ACoS for Amazon ads?
There is no universal answer. A good ACoS is one that sits below your break-even ACoS for a given product. Break-even ACoS is your profit margin expressed as a percentage.
If your product has a 35% margin after all fees and COGS, then any ACoS below 35% means you’re making money on ad-attributed sales. Most established sellers target 15–25% for profitable products, but the right number for your account depends entirely on your margin structure and whether you are optimizing for profitability or for growth and organic rank building.
What is the difference between ACoS and TACoS, and which one should I be watching?
ACoS measures ad spend as a percentage of ad-attributed revenue only. TACoS (Total Advertising Cost of Sale) measures ad spend as a percentage of total revenue — organic plus paid.
TACoS is the more meaningful long-run metric because it captures whether your paid advertising is helping or hurting your overall business health.
An ACoS that rises slightly while TACoS falls is actually a good signal — your ads are generating organic momentum that compounds beyond the direct attribution window. An ACoS that holds steady while TACoS rises means your organic business is weakening and your ads are doing increasingly more work to hold total revenue together.
My ACoS went up but spend and revenue both increased. Is that still a problem?
Not automatically; this is usually a volume and efficiency trade-off.
When you expand into broader match types, increase bids to capture more top-of-search inventory, or scale into new keywords, both spend and revenue can rise together. But the incremental revenue often comes at a lower efficiency than your existing base, so ACoS rises as you scale. Whether this is a problem depends entirely on whether the ACoS at the new volume still sits below your break-even threshold.
If it does, you have a healthy scaling trade-off. If it does not, you are buying revenue at a loss.
How do I know if an ACoS spike is from a campaign problem or a listing problem?
CVR is the clearest signal. If conversion rate dropped and impressions or clicks held flat, the problem is almost certainly on the product side rather than in the campaign.
Campaign problems tend to show up as traffic problems first — fewer impressions, lower CTR, rising CPC. Listing problems show up as conversion problems — the traffic arrives, the click happens, but the purchase does not.
Before touching any bids, check your listing for review count and rating, buy box ownership, current price relative to the top results, and whether stock levels are creating extended delivery estimates.
Does lowering bids fix a high ACoS?
Sometimes, but only for a specific cause. If ACoS is high because CPC rose due to auction pressure and CVR did not improve to compensate, lowering bids can reduce spend and improve ACoS efficiency.
If ACoS is high because CVR dropped due to a listing problem, lowering bids only reduces traffic volume going to a listing that still isn’t converting — your ACoS may barely move, and your total sales will fall. If ACoS is high because of a demand or seasonality shift, lowering bids reduces visibility without addressing the market condition driving the problem. Identify the cause first. Then choose the intervention.
Can a competitor’s actions cause my ACoS to spike without me changing anything?
Yes, and it is more common than most people realize. If a competitor launches a deep promotion, drops their price significantly, or launches a product that directly competes with your listing, your CVR will fall even if every element of your campaign is unchanged. Shoppers are now choosing them over you.
Similarly, if a major competitor enters your keyword auction at higher bids, your CPC will rise and your ACoS will follow unless your CVR improves proportionally. These are market-level signals. They require you to understand what is happening in the competitive landscape, not just in your campaign structure.
Amazon does not surface these relationships in native reporting, which is exactly why a structured diagnostic process matters more than a reactive dashboard check.
What is the relationship between ACoS and ROAS?
ACoS and ROAS are mathematical inverses of each other.
ACoS = Ad Spend ÷ Ad Revenue × 100
ROAS = Ad Revenue ÷ Ad Spend.
So a 25% ACoS is equivalent to a 4x ROAS. A 50% ACoS is equivalent to a 2x ROAS.
Amazon’s native interface uses ACoS as the primary efficiency metric, while some third-party tools and Google Ads practitioners tend to reference ROAS.
Either way, you are measuring the same relationship between spend and revenue — just from different directions. For Amazon, ACoS is the industry-standard way to talk about it, so it is worth being comfortable converting between the two when you are comparing platform benchmarks.
How long should I wait before acting on an ACoS change?
At minimum, you should wait until the data window you are analyzing has closed its attribution period — 7 days for Sponsored Products, 14 days for Sponsored Brands. Acting on week-one data in a 7-day attribution window means you are making decisions on incomplete conversion numbers.
As a practical rule: daily ACoS fluctuations of 10–15% are common and often noise. A sustained move over two or more weeks, across multiple campaign types, that holds after the attribution window closes is a signal worth investigating.
You can use Optmyzr’s Performance Comparison with settled date ranges rather than the rolling default windows in Amazon’s native UI.
