ACoS vs ROAS: Which Metric to Optimize For?

ACoS vs ROAS: the short answer

ACoS and ROAS are two views of the same relationship between Amazon ad spend and attributed sales. ACoS tells you how much of attributed sales revenue is being spent on advertising. ROAS tells you how many dollars of attributed sales are generated for each advertising dollar.

The formulas are:

• ACoS = Ad Spend ÷ Ad Sales × 100
• ROAS = Ad Sales ÷ Ad Spend
• ROAS = 1 ÷ ACoS when ACoS is expressed as a decimal.

For example, if you spend $1,000 and generate $4,000 in attributed sales:

• ACoS = 25%
• ROAS = 4.0

Neither metric is inherently better. The useful question is which view makes the business decision clearer.

If the business is managing an advertising-cost ceiling, ACoS is usually easier to communicate. If the business is evaluating revenue generated per advertising dollar or comparing budget opportunities, ROAS can be easier to interpret.

The important point is that the target should come from product economics and campaign purpose, not from an industry benchmark.

Amazon's advertising guidance also treats ACoS and ROAS as inverse measures and notes that there is no universal "good" ACoS. Amazon's ACoS guide provides the underlying definitions.

ACoS vs ROAS: what each metric tells an operator

The difference is mainly how you frame the decision.

If you are trying to keep advertising below a certain percentage of sales, ACoS is intuitive.

If you are deciding where another advertising dollar could go, ROAS can be easier to interpret.

But neither metric tells you whether the next dollar will be profitable.

ACoS and ROAS are inverse metrics, not competing strategies

A 25% ACoS is a 4.0 ROAS.

A 20% ACoS is a 5.0 ROAS.

A 40% ACoS is a 2.5 ROAS.

There is no strategic advantage in choosing one metric because the underlying advertising efficiency is different. The same spend and sales produce both numbers.

The strategic difference comes from what you do with the information.

An operator might say:

"Keep this campaign around 25% ACoS."

Another might say:

"Maintain at least 4.0 ROAS."

If both refer to the same attributed sales relationship, they are expressing essentially the same efficiency constraint.

The real strategy sits behind the target.

The mistake: optimizing the metric instead of the business

This is where many Amazon PPC accounts go wrong.

An operator sees ACoS rise from 25% to 32% and immediately cuts bids.

Or ROAS falls from 4.0 to 3.1 and the campaign gets throttled.

That can improve the reported metric while making the business worse.

The metric does not tell you why performance changed.

Higher ACoS or lower ROAS could come from:

• Higher CPC
• Lower conversion rate
• More expensive placements
• More discovery traffic
• A deliberate push into non-brand terms
• A new product with limited history
• A price or promotion change
• A shift in traffic mix

The first job is diagnosis. The second is optimization.

That is the same principle used in a proper Amazon PPC audit: identify what changed, determine whether the change is useful or wasteful, then decide what to adjust.

When ACoS is the better operating metric

ACoS is particularly useful when the business thinks in terms of advertising cost as a percentage of revenue.

Suppose a product generates $10,000 in attributed sales:

• 20% ACoS = $2,000 ad spend
• 30% ACoS = $3,000 ad spend
• 40% ACoS = $4,000 ad spend

That makes ACoS useful as an operating guardrail.

For example:

"Keep non-brand advertising around 25% ACoS while protecting sales volume."

That is a reasonable operating constraint.

The mistake is turning it into a rigid target for every campaign.

ACoS is often easier to use when:

1. 1The business has a defined advertising-cost ceiling.
2. 2Product margins are relatively stable.
3. 3The main objective is profitable sales efficiency.
4. 4Management wants advertising expressed as a percentage of revenue.
5. 5The campaign is mature enough that efficiency is more important than aggressive discovery.

Break-even ACoS matters more than an industry benchmark

A target ACoS should start with product economics.

Suppose:

• Selling price = $60
• Contribution before advertising = $18

The theoretical break-even ACoS is:

$18 ÷ $60 = 30%

At 20% ACoS, advertising consumes $12 of the sale.

At 30%, it consumes $18.

At 35%, it consumes $21, exceeding that simplified contribution.

This is why copying another seller's "target ACoS" is weak operating practice.

Your break-even point depends on your economics.

And break-even is not the same as the target you should actually run.

A growth campaign may deliberately operate above break-even for a period. A mature brand-defense campaign may require a different threshold because some attributed sales might have happened without the ad.

For more detail, see how to lower ACoS on Amazon.

When ROAS is the better operating metric

ROAS reverses the same equation.

If a campaign spends $2,000 and generates $8,000 in attributed sales:

ROAS = 4.0

Every $1 of advertising spend generated $4 in attributed sales.

ROAS becomes useful when you want a simple revenue-per-dollar view.

For example:

• Campaign A: $500 spend, $2,000 sales, 4.0 ROAS
• Campaign B: $5,000 spend, $15,000 sales, 3.0 ROAS

Campaign A has better reported efficiency.

But that does not automatically mean Campaign A should receive the next $5,000 of budget.

Campaign B is already producing substantially more sales and may have more room to grow.

This is one of the most important operator distinctions:

Efficiency and scale are not the same thing.

Amazon's advertising guidance also emphasizes that the campaign with the highest ROAS is not automatically the right campaign for every business objective. Amazon's advertising budget guidance discusses how budget decisions should reflect campaign goals.

ACoS vs ROAS vs TACoS

ACoS and ROAS are focused on attributed advertising sales.

TACoS puts advertising spend against total sales.

The formulas are:

• ACoS = Ad Spend ÷ Ad Sales
• ROAS = Ad Sales ÷ Ad Spend
• TACoS = Ad Spend ÷ Total Sales

TACoS can help answer a different question:

How much of the entire Amazon revenue base am I spending on advertising?

Consider this example.

Month 1

• Ad spend: $10,000
• Ad sales: $40,000
• Total sales: $80,000
• ACoS: 25%
• ROAS: 4.0
• TACoS: 12.5%

Month 2

• Ad spend: $10,000
• Ad sales: $45,000
• Total sales: $100,000
• ACoS: 22.2%
• ROAS: 4.5
• TACoS: 10%

The advertising metrics improved, and total sales grew.

That is a much more useful business outcome than simply saying "ROAS went up."

But TACoS also has limits. It does not prove that advertising caused the organic sales increase. For that, you need stronger incrementality analysis.

ACoS and ROAS do not tell you profit by themselves

Both metrics use attributed ad sales.

They do not subtract:

• Product cost
• Amazon fees
• Fulfillment
• Discounts
• Returns
• Other variable costs

So:

20% ACoS does not mean 20% profit.

And:

5.0 ROAS does not mean a 5x return on profit.

Suppose a product sells for $50 and contributes $15 before advertising.

At 20% ACoS:

• Ad spend = $10
• Contribution before advertising = $15
• Contribution after advertising = $5

At 30% ACoS:

• Ad spend = $15
• Contribution before advertising = $15
• Contribution after advertising = $0

That is why product economics should determine the acceptable advertising range.

The same ACoS can mean very different things

Suppose two campaigns both have 30% ACoS.

Campaign 1: Brand defense

The campaign targets shoppers already searching for the brand.

It may have excellent conversion and stable sales.

But some of those customers may have purchased anyway.

Campaign 2: Non-brand acquisition

The campaign targets generic category searches.

It may have lower conversion and higher acquisition cost.

But it could be introducing the brand to shoppers who were not already looking for it.

Treating both campaigns with the same target can lead to poor decisions.

This is why experienced operators segment targets by campaign role, not only by ACoS or ROAS.

For Sponsored Brands, Amazon also provides new-to-brand metrics, which can add context when the campaign is intended to acquire new customers rather than simply capture existing branded demand. Amazon's Sponsored Brands guidance explains these additional measurement options.

Campaign role should influence the metric you emphasize

This does not mean every campaign needs a different ACoS target.

It means the same number can have different business meanings depending on why the campaign exists.

The real question: where should the next dollar go?

This is where the conversation moves beyond ACoS vs ROAS.

Suppose:

Campaign A

• Spend: $1,000
• Sales: $5,000
• ROAS: 5.0
• ACoS: 20%

Campaign B

• Spend: $10,000
• Sales: $40,000
• ROAS: 4.0
• ACoS: 25%

Campaign A looks better on efficiency.

But before moving budget from B to A, ask:

• Is A already close to its available demand?
• Can A absorb another $1,000?
• What happens to CPC as spend increases?
• Is B constrained by budget?
• Does B have profitable search-term headroom?
• Is B driving incremental non-brand demand?
• Are the products behind both campaigns equally profitable?

The correct question is not:

"Which campaign has the highest ROAS?"

It is:

"Where is the next advertising dollar most likely to produce useful incremental sales at an acceptable contribution?"

That is a much better budget-allocation question.

What I optimize first as a PPC operator

I would not start with:

"How do I get ACoS down?"

I would start with:

"What is the business trying to achieve, and where is the current advertising spend helping or hurting that objective?"

Then I work through the account in this order.

1. Product economics

Know the contribution available before advertising.

Without this, ACoS targets are mostly arbitrary.

2. Campaign role

Separate:

• Brand defense
• Non-brand acquisition
• Product discovery
• Competitor targeting
• Category expansion
• Audience activity where applicable

A single target across all of them is often too crude.

3. Search terms and targets

Find where spend is producing useful sales and where it is simply consuming budget.

Relevant but expensive traffic may need a bid change.

Irrelevant traffic may need negative targeting.

Those are different decisions.

4. Conversion rate

If clicks are healthy but orders are weak, reducing bids may hide the symptom without fixing the cause.

Check the product detail page, price, reviews, offer, images, and competitive position.

5. Budget allocation

A campaign can have excellent ROAS and still be a poor place for the next dollar if it has limited incremental opportunity.

Conversely, a campaign with lower ROAS can deserve more budget if it has meaningful headroom and acceptable contribution economics.

6. Incremental performance

Ask whether additional spend is creating additional business or simply receiving credit for sales that were already likely to happen.

This is especially important with branded traffic and mature products.

7. Blended business effect

Look beyond the advertising dashboard.

Review:

• Ad sales
• Organic sales
• Total sales
• TACoS
• Contribution
• New-to-brand metrics where relevant

The objective is not to produce a beautiful PPC dashboard.

The objective is to improve the economics of the Amazon business.

The operator's dashboard should show both

There is no reason to choose one metric and hide the other.

A practical Amazon PPC view can include:

• Spend
• Ad sales
• ACoS
• ROAS
• Orders
• Conversion rate
• CPC
• Clicks
• Organic sales
• Total sales
• TACoS
• New-to-brand metrics where relevant
• Contribution or profit estimate where available

Amazon provides ACoS and ROAS in its advertising reporting, while its campaign measurement guidance recommends looking at multiple performance signals rather than relying on a single number. Amazon's campaign measurement guidance covers this broader approach.

The metric should support the decision, not replace it.

ACoS vs ROAS conversion table

Because the metrics are inverses, you can translate between them quickly:

Use this table to translate reports, not as a universal performance benchmark.

A better way to set your Amazon PPC target

Use this sequence:

1. Calculate contribution before advertising

How much money is available from each sale before ad spend?

2. Establish the economic ceiling

What ACoS can the product support before contribution is exhausted?

3. Define the campaign role

Is the campaign defending existing demand, acquiring new customers, discovering search terms, or expanding the category?

4. Set the operating target

Use ACoS or ROAS as the practical control metric.

5. Add growth constraints

Do not sacrifice valuable incremental sales simply to make the metric look better.

6. Review marginal performance

Ask what happened when you added or removed the last meaningful amount of spend.

7. Check the blended result

Review TACoS, total sales, organic sales, contribution, and relevant customer-acquisition signals.

That is much closer to how budget decisions should actually be made.

The biggest mistake: optimizing the metric instead of the business

If you remember one thing from this article, make it this:

Do not optimize ACoS or ROAS in isolation. Optimize the economics behind them.

ACoS is useful for understanding advertising cost relative to attributed sales.

ROAS is useful for understanding attributed sales generated per advertising dollar.

TACoS helps put advertising spend against the total Amazon revenue base.

None of these metrics should replace operator judgment.

They are measurement tools.

The real decision is whether advertising is producing the right sales at an acceptable economic return, whether those sales are genuinely incremental, and where the next dollar of spend has the strongest case.

For a broader operating framework, see the complete guide to Amazon PPC optimization.

About the author

Yogendra Kashyap

Amazon Ads operators

Yogendra Kashyap is the Founder of SellerRoot and an Amazon Ads expert with 9+ years of experience helping brands grow through data-driven advertising. His expertise spans Amazon PPC, campaign optimization, search term analysis, and marketplace growth. Together with the SellerRoot team, he is building AI-powered tools for Amazon advertisers while sharing practical, experience-backed insights to help brands improve profitability and scale on Amazon.