A 4:1 return on ad spend can be a strong quarter for one store and a
slow leak for another, and the ad account reports both the same way. A
good ROAS is any ROAS above your break-even ROAS, which is 1 divided by
your contribution margin; the 2:1 to 4:1 range you see quoted everywhere
is a rule of thumb, not a target.
This guide shows the break-even math, how to turn it into a target,
which 2026 ROAS benchmarks actually have a credible source behind them,
and what changes when your revenue lives in a CRM instead of a checkout.
What Is a Good ROAS?
**A good ROAS is any return on ad spend above your break-even ROAS,
measured on revenue you can verify.** ROAS is ad-attributed revenue
divided by ad spend, and the 2:1 to 4:1 range is only a starting point.
At a 40% contribution margin, you break even at 2.5x, and anything
higher goes toward overhead and profit; at a 25% margin, you need 4x
just to stand still.
Synter's own [ROAS
calculator](https://syntermedia.ai/tools/roas-calculator)
labels 400% (4:1) and above as excellent, but 4:1 only clears
variable-cost break-even above a 25% contribution margin. Excellence
depends on what remains for overhead and profit. A brand with a 40%
margin keeps 15 cents of every revenue dollar after ads at 4:1, while a
business on grocery-like margins sits at or below break-even on the same
number.
ROAS as a Ratio, a Multiple and a Percentage
The same result gets written three ways:
Ratio Multiple Percentage ACoS (spend ÷
revenue)
- ----------------- ----------------- ----------------- -----------------
2:1 2x 200% 50%
3:1 3x 300% 33%
4:1 4x 400% 25%
5:1 5x 500% 20%
To calculate ROAS, divide attributed revenue by ad spend: \$4,000 of
revenue on \$1,000 of spend is 4:1, or \$4 back for every \$1. Google
Ads expresses targets as percentages, so a 4:1 goal is a 400% Target
ROAS. Amazon Ads talks in ACoS, which is ad spend divided by ad revenue,
so ROAS is simply its inverse.
How to Calculate Your Break-Even ROAS
**Break-even ROAS, the minimum ROAS a campaign needs to pay for itself,
is 1 ÷ contribution margin.** Pre-ad contribution margin is revenue
minus non-ad costs that rise with every order, and it is the standard
basis for [break-even
analysis](https://www.investopedia.com/terms/c/contributionmargin.asp).
For ad math, that means cost of goods, shipping and packing, payment
fees, and expected refunds. Fixed rent and salaries belong in the
cushion you add later; labor that varies with orders belongs in
contribution margin. Net profit margin is the wrong input, because it
already has overhead and past ad spend taken out: Retail (General) runs
a 5.61% net margin in Aswath Damodaran's January 2026 industry data,
which would imply an absurd 17.8x break-even.
Amazon frames the same rule from the other side: to stay profitable,
[your ACoS needs to be lower than your profit
margin](https://advertising.amazon.com/library/guides/acos-advertising-cost-of-sales).
Break-even ACoS equals your margin, and break-even ROAS is its inverse.
Contribution margin Break-even ROAS Break-even ACoS
- ----------------------- ----------------------- -----------------------
20% 5.00x 20%
30% 3.33x 30%
40% 2.50x 40%
50% 2.00x 50%
60% 1.67x 60%
Here is a worked example for an \$80 average order:
- Cost of goods: \$30
- Shipping and packing: \$9
- Payment processing at 3%: \$2.40
- Returns and refunds allowance: \$4
That leaves \$34.60 of contribution per order, a 43.25% margin, so this
store breaks even at 2.31x. Every campaign below that line loses money
on each sale it reports, and every campaign above it adds toward
overhead and profit.
From Break-Even to Target ROAS
Break-even covers the order, not the team, the software, or the profit
you are in business to make, so **your target ROAS is break-even plus a
cushion**. Decide what share of revenue should be left after variable
costs and ad spend, then solve for it:
**Target ROAS = 1 ÷ (contribution margin - share kept for overhead and
profit)**
That share must be nonnegative and smaller than your pre-ad margin. For
the \$80 store above, keeping 10% after variable costs and ads means 1 ÷
(0.4325 - 0.10), or roughly a 3.0x target. Keeping 15% at a 40% margin
means 1 ÷ 0.25, which lands exactly on the 4:1 rule of thumb.
Prospecting can run below that line on purpose. If a new customer's
12-month contribution is twice the contribution from their first order,
a first-order ROAS of 1.25x breaks even at a 40% margin, because the
repeat orders pay back the customer acquisition cost. Retargeting and
branded search should be held to a higher target, because they reach
people who already know you and some of that revenue would have arrived
anyway.
The target also matters more in Google Ads than it used to. Since 17
August 2026, budget-limited campaigns on Target ROAS or Target CPA that
beat their target now trend toward the target you set. We covered the
[setpoint change and how to reset
targets](https://syntermedia.ai/blog/google-target-cpa-bidding-change-august-2026)
when it shipped: a target set loosely below your real ROAS is now a
number Google will spend down to. Google also warns that a target set
too high may limit traffic, so the target is a profit decision with a
volume cost attached in both directions.
ROAS Benchmarks by Platform
We found no current ROAS benchmarks published by Meta, LinkedIn,
Microsoft, or TikTok, and the figures that do circulate for those
platforms mostly come from ad-tech vendors' own customer panels. Treat
every row as directional.
Platform Published benchmark (source, date) What the reported ROAS counts Compute instead
- ----------------- ------------------------------------------------------------------------------------------- -------------------------------------------------------------------------------------------------------------------------------- -----------------
Google Ads No ROAS benchmark. Google's economic impact model assumes \$2 of profit per \$1 of Conv. value/cost uses included conversion values. Click windows default to 30 days; check each action ([Conversion Conv. value/cost
(Search, Google spend, a 2009 estimate ([Google, 2025 windows). View-through appears separately and in All conversions, by campaign type,
Shopping, PMax) methodology]{.underline}](https://economicimpact.google/methodology/)) not Conversions (Google Ads Help) with brand and
non-brand split
Meta Ads None published Standard attribution includes link-click, view-through and engage-through conversions within your selected ad-set windows; check Break-even on
current settings (Meta Business Help; [Meta measurement new-customer
update](https://about.fb.com/ltam/news/2026/03/simplificando-la-medicion-de-anuncios-para-un-mundo-social-first/)) revenue; compare
with Meta's
incremental
attribution model
Amazon Sponsored No Sponsored Products benchmark. Amazon's guide calls 2:1 slightly above the general Amazon-attributed sales; Amazon's guides frame efficiency as ACoS Break-even ACoS =
Products industry average and suggests 3 to 4, citing Nielsen's 2016 cross-media CPG research your margin
([Amazon
Ads](https://advertising.amazon.com/library/guides/return-on-ad-spend-roas))
LinkedIn Ads None published Revenue Attribution Report: revenue won ÷ LinkedIn spend, 180-day default lookback ([LinkedIn Pipeline and
Help](https://www.linkedin.com/help/lms/answer/a1367472)) closed-won ROAS
over your sales
cycle
Microsoft Ads None published The conversion window you configure, up to 90 days ([Microsoft Same as Google,
Learn](https://learn.microsoft.com/en-us/advertising/campaign-management-service/conversiongoal?view=bingads-13)) on a matching
window
TikTok Ads None published A click-through window of 1 to 28 days and view-through of up to 7 days, chosen per ad group ([TikTok Business Break-even on
Help](https://ads.tiktok.com/help/article/attribution-settings-at-the-ad-group-level?lang=en)) new-customer
revenue, judged
on a blended view
Google's \$2 figure is profit, not ROAS. The "\$2 per \$1" number
sometimes quoted as a 2:1 ROAS for Google Ads is an assumption about
advertiser profit from Hal Varian's 2009 work, not a measured ROAS.
What is a good ROAS on Amazon? Amazon's guide suggests 3 to 4 and
calls 2:1 slightly above the industry average. Its footnote cites
Nielsen's 2016 cross-media CPG research, not Sponsored Products data.
Treat it as general context; set your target from break-even ACoS, which
equals your margin.
ROAS Benchmarks by Industry
We could not find independent, dated ROAS benchmarks by industry for
- Inverting published sector gross margins gives illustrative
break-even ratios, not observed ROAS benchmarks. Your own product mix
and cost categories will move them. The table uses Aswath Damodaran's
[margins by industry
dataset](https://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/margin.html)
at NYU Stern, built on January 2026 data for US firms.
Industry (Damodaran Gross margin, Jan 2026 Illustrative break-even
category) ROAS (1 ÷ gross margin)
- ----------------------- ----------------------- -----------------------
Apparel 56.88% 1.76x
Household Products 51.04% 1.96x
Electronics (Consumer & 38.77% 2.58x
Office)
Retail (Special Lines) 35.30% 2.83x
Retail (General) 33.18% 3.01x
Furn/Home Furnishings 30.28% 3.30x
Retail (Grocery and 26.31% 3.80x
Food)
Software (System & 71.72% 1.39x
Application)
All US firms (Total 37.76% 2.65x
Market)
These are floors, not targets. Gross margin only subtracts cost of
goods sold. Once you also take out the other costs that rise with each
order, such as payment fees, refunds, and, for many companies, shipping,
your contribution margin is lower, and your real break-even is higher
than the number in the table.
What is a good ROAS for ecommerce? A retailer at the sector's 33%
gross margin loses money below about 3x before any other per-order cost
is counted, so a general retailer needs a target above 3x and should
check whether 4x leaves any profit at all. Nielsen Catalina Solutions'
[study of 1,400+ CPG
campaigns](https://www.nielsen.com/insights/2016/benchmarking-return-on-ad-spend-media-type-brand-size-matter/)
compared exposed and unexposed households and found the top category,
baby products, generated a return on spend of \$3.71, with pets second
at \$3.06. That data is from 2016 and covers in-store CPG sales, so read
it as a reality check rather than a 2026 benchmark.
Software has a low illustrative floor, but the category covers system
and application software broadly, not only B2B SaaS. A campaign judged
only on first-month revenue can show ROAS below 1x and still pay back
over the contract term.
What a Good ROAS Looks Like for B2B and Lead Gen
For lead gen, ROAS does not exist until [CRM revenue is tied back to the
ad](https://syntermedia.ai/blog/b2b-marketing-attribution). Until then,
the "ROAS" in the platform is whatever value you assigned to a form
fill. Assign \$50 to each of 120 leads from a \$40,000 LinkedIn quarter,
and the platform reports 0.15x, which describes your conversion
settings, not your campaign.
**[Pipeline
ROAS](https://syntermedia.ai/blog/pipeline-attribution-reporting) is the
early read.** It divides the value of opportunities created by spend, so
\$360,000 of pipeline on that same \$40,000 is 9.0x. It shows up long
before deals close, which makes it useful for steering, but it also
counts deals that never will.
Closed-won ROAS is the one to set targets on. It divides revenue won
by spend. At a 25% win rate, that pipeline becomes \$90,000, or 2.25x on
first-year contract value, above the illustrative 1.39x software line,
but per-deal costs such as commissions and onboarding raise real
break-even; check your own contribution margin. Multi-year contracts and
expansion revenue push the real return higher still.
The platforms support this, and the settings matter. LinkedIn's Revenue
Attribution Report defines ROAS as revenue won divided by LinkedIn ad
spend and counts influence over a 180-day default lookback. Google Ads
lets you import offline conversions to measure what happens after a
click or call, and its click-through window can stretch to 90 days for
some conversion sources. Deals that take longer than those windows still
fall outside them, so judge long-cycle channels on CRM data rather than
in-platform reports.
This is the problem Synter's
is built around. It treats CRM closed-won as the truth set, pulls deal
and revenue data from HubSpot and Salesforce, and compares first-click,
last-click, linear, time-decay, and position-based models side by side.
Offline conversion upload is included on every plan, which lets
closed-won values flow back to the ad platforms, while CRM sync and
multi-touch attribution sit on the Scale and Custom plans. It is a way
to get a verifiable ROAS across platforms, not a source of benchmarks.
Why Your Reported ROAS Might Be Wrong
A correct target still fails if the revenue in the numerator is wrong.
As our [CMO
checklist](https://syntermedia.ai/blog/cmo-checklist-ai-ad-agents)
puts it, "Platform-reported ROAS is each platform grading its own
homework, and every platform claims the same conversion."
Platforms double count. Google counts click-through conversions for
30 days by default, while Meta can count purchases within the ad set's
selected click or view-through window. A customer who clicked a search
ad and later saw an Instagram ad can be claimed by both, so a store with
\$100,000 of total revenue can, for example, see \$125,000 claimed
across two dashboards.
Windows differ, so the numbers do not compare. [Different
attribution
models](https://syntermedia.ai/blog/marketing-attribution-models) assign
credit differently. Putting a 30-day click number next to a 7-day click
number compares the settings as much as the channels.
Blended revenue flatters prospecting. Platform ROAS mixes new and
returning customers, and returning buyers lift the number without
growing the business, so set separate targets for prospecting and
retargeting.
Returns rarely reach the ad account. Platforms count the order, not
the refund, unless you upload restatements or retractions afterward,
which Google Ads supports through [conversion
adjustments](https://support.google.com/google-ads/answer/7686280?hl=en).
Retailers surveyed by NRF expected 19.3% of online sales to be returned
in 2025 ([NRF, October
2025](https://nrf.com/media-center/press-releases/consumers-expected-to-return-nearly-850-billion-in-merchandise-in-2025)).
At that rate, a reported 4.0x is closer to 3.2x in revenue you keep.
Attributed is not the same as caused. eBay's paid search field
experiments found returns were a fraction of non-experimental estimates
and that brand-keyword ads had no measurable short-term benefit
([Blake, Nosko and Tadelis,
NBER](https://www.nber.org/papers/w20171)).
How to Improve ROAS Without Killing Growth
**For click-based ecommerce, ROAS = conversion rate × average order
value ÷ cost per click, with conversion rate measured per paid click.
Every commercial lever works through one of those three terms.** At a
\$1.20 CPC, a 3% conversion rate, and an \$80 average order, ROAS is
- 2.0x. Lift conversion rate to 3.6%, and it becomes 2.4x; lift order
value to \$96 instead, and you get the same 2.4x; do both, and it
reaches 2.88x.
Work through the levers in this order:
- Fix measurement first. [Server-side conversion
tracking](https://syntermedia.ai/blog/server-side-conversion-tracking)
and CRM revenue sent back to the platforms give bidding the right
numbers to optimize. This changes measurement, so reset your
baseline afterward.
- Cut search-term waste. Non-converting queries add spend without
adding revenue.
- Raise landing page conversion rate. It sits in the numerator, so
at the same CPC a 20% lift raises ROAS by 20%.
- Raise average order value with bundles and free-shipping
thresholds.
- Refresh creative before fatigue shows up as falling
click-through.
- Move budget toward campaigns above target, judged on the last
dollar spent rather than the average.
The trap is treating ROAS as the thing to maximize. At a 40% margin,
\$10,000 of spend at 5.0x returns \$50,000 of revenue and \$10,000 of
contribution after ads. Push spend to \$30,000 and ROAS drops to 3.5x,
yet revenue reaches \$105,000 and contribution rises to \$12,000. Lower
ROAS on more spend can be the better business, as long as it stays above
target.
Synter handles this kind of reallocation through its MCP tools. Our
optimize_budget tool allocates spend across channels using
diminishing-returns modeling, and our forecast_campaign projects ROAS
and CPA for the next 7 to 30 days with confidence intervals. Spend
changes can be held behind a human approval gate, and the execute tool
rejects any action that isn't registered. Our [cross-channel budget
allocation
guide](https://syntermedia.ai/blog/cross-channel-ad-budget-allocation-guide)
covers the allocation models.
Frequently Asked Questions
Is a 3 ROAS good?
It is if your contribution margin is above 33%, where 3x is break-even.
At a 40% margin, 3x leaves 20 cents per dollar of spend for overhead and
profit. At 30%, the same ROAS loses 10 cents on every dollar.
Is 2 ROAS profitable?
It leaves contribution toward overhead and profit only above a 50%
pre-ad margin. At 2x, ads cost half of the revenue they bring in, which
works for high-margin products, subscription business models with strong
repeat revenue, or prospecting you expect to recover on later orders.
What ROAS is 25% ACoS?
25% ACoS is a 4x ROAS (400%), because ACoS is spend divided by revenue
and ROAS is the inverse. It breaks even at a 25% margin.
Is a 2.2 ROAS good?
It depends on your margin, because 2.2x breaks even at a 45.5%
contribution margin. Under that margin, it loses money, and over it the
campaign leaves contribution toward overhead and profit. The \$80 store
above, at 43.25%, loses about 5 cents per dollar of spend at 2.2x, while
a brand at a 60% margin earns 32 cents.
Is a 7 ROAS good?
It leaves contribution toward overhead and profit above a 14.3% pre-ad
margin. It can also mean you are underinvesting or mostly buying branded
and retargeting traffic, so test more budget before celebrating.
What is a good ROAS for Google Ads?
Any ROAS above your break-even, measured on conversion values you trust.
Google does not publish a ROAS benchmark, and its own guidance is to set
Target ROAS from your business goals with your historical ROAS as a
reference. Search and Shopping campaigns also need at least 15
conversions in the past 30 days before Target ROAS bidding is available.
What is the difference between ROAS and ROI?
ROAS measures revenue per ad dollar, while ROI measures profit on the
total investment.
Conclusion
A good ROAS is a number you calculate from your margin, not one you look
up. Start with break-even (1 ÷ contribution margin), add a cushion for
overhead and profit to get a target, treat benchmarks as context, and
check that the revenue behind the number is real before you act on it.
If your revenue lands in a CRM, the next step is putting closed-won data
next to ad spend on every platform, then setting targets per campaign.
discuss setting ROAS targets on closed-won revenue.