What is ROAS (Return on Ad Spend)?
Also known as: Return on ad spend, Ad ROI
What is ROAS?
ROAS, short for Return on Ad Spend, is the revenue an advertising campaign generates for every dollar spent on that campaign.
It is the single most-watched metric in performance marketing. Every Meta, Google, and TikTok dashboard puts ROAS at the top. Every paid media review starts with it.
The formula is simple:
ROAS = Revenue from ads / Ad spendA campaign that spends $10,000 and produces $40,000 in revenue has a 4.0 ROAS. Sometimes ROAS is shown as a multiplier (4.0x). Sometimes as a percentage (400 percent). The math is identical.
ROAS answers one question fast: did this ad spend produce more revenue than it consumed?
ROAS formula and worked examples
The formula has two inputs. Both need to be defined carefully.
Revenue from ads is the gross revenue attributed to the campaign by the ad platform or your analytics stack. It does not subtract product cost, shipping, refunds, or platform fees.
Ad spend is the total amount paid to the ad network. It usually does not include creative production, agency fees, or tooling costs.
Three quick examples:
| Spend | Revenue | ROAS | Read as |
|---|---|---|---|
| $5,000 | $25,000 | 5.0 | $5 returned per $1 spent |
| $12,000 | $18,000 | 1.5 | $1.50 returned per $1 spent |
| $8,000 | $4,000 | 0.5 | $0.50 returned per $1 spent (loss) |
Per Google Ads' official ROAS documentation, the platform calculates Conv. value / cost using the conversion values you set in your account. If conversion values are missing or wrong, the reported ROAS is wrong.
ROAS vs ROI vs CPA
The three metrics measure different things. Confusing them is the most common reporting mistake in paid media.
| Metric | Formula | What it answers |
|---|---|---|
| ROAS | Revenue / Ad spend | Did ads return more revenue than they cost? |
| ROI | (Profit, Investment) / Investment | Did the campaign actually make money after all costs? |
| CPA | Ad spend / Conversions | What did each new customer cost to acquire? |
ROAS is the easiest to calculate. ROI is the most honest. CPA is the most useful for budget pacing.
A campaign can hit a 4.0 ROAS and still lose money if the gross margin is below 25 percent. A campaign with a low ROAS can still be profitable if the lifetime value of each new customer is high.
Smart paid teams report all three on every campaign review.
What is a "good" ROAS?
There is no universal answer. A "good" ROAS depends on margin, business model, and customer lifetime value.
Three reference points by industry:
- Direct-to-consumer ecommerce: 3.0 to 5.0 break-even, 6.0+ profitable. Margins after COGS and shipping rarely exceed 35 percent.
- SaaS and digital products: 1.5 to 3.0 break-even on first purchase, much higher when LTV is factored in. Subscription revenue compounds.
- Lead generation (B2B, finance, insurance): ROAS is often replaced by cost per qualified lead because revenue lags weeks or months behind the click.
Per Statista's 2024 ROAS benchmark report, Meta Ads ROAS varies from 1.2 in highly competitive verticals (financial services) to 4.5+ in retail with strong creative. The platform median sits at roughly 2.8.
The right benchmark for your account is your own break-even. Calculate it once:
Break-even ROAS = 1 / Contribution marginA brand with 40 percent contribution margin breaks even at 2.5 ROAS. Anything above is profit. Anything below is a subsidy.
Why ROAS alone is misleading
ROAS is loud. It is also incomplete.
Three structural problems:
1. It ignores margin. A 5.0 ROAS on a 10 percent margin product loses money. A 1.8 ROAS on a 70 percent margin product prints cash. The metric does not care.
2. It ignores LTV. First-purchase ROAS is a snapshot. If a customer buys five more times over 18 months, the true return is 6x what the first-click report shows. Subscription, replenishment, and high-NPS brands underprice their ads when they optimize on first-purchase ROAS only.
3. It depends on attribution. Per Meta's Conversion Lift documentation, platform-reported conversions can over-credit ad spend because the attribution window gives Meta credit for conversions that would have happened anyway. Lift studies regularly show 20 to 40 percent of attributed conversions are incremental, not the 100 percent the dashboard implies.
The fix is not to drop ROAS. It is to triangulate. Pair platform-reported ROAS with blended ROAS, CAC payback, and incremental lift testing.
Real-world example with numbers
A direct-to-consumer coffee brand runs a $50,000 monthly Meta budget across three campaigns.
| Campaign | Spend | Reported revenue | Reported ROAS |
|---|---|---|---|
| Prospecting (cold) | $30,000 | $66,000 | 2.2 |
| Retargeting | $12,000 | $84,000 | 7.0 |
| Brand search | $8,000 | $72,000 | 9.0 |
| Total | $50,000 | $222,000 | 4.4 |
The dashboard shows a healthy 4.4 ROAS. Then the finance team runs the real numbers.
Gross margin on coffee is 32 percent. Of $222,000 in revenue, gross profit is $71,040. Subtract the $50,000 in ad spend, and the campaign cleared $21,040 in contribution.
Then they run a holdout test on retargeting and brand search. The lift study shows 60 percent of those "conversions" would have happened without the ads, the customers were already coming back. True incremental revenue drops to roughly $133,000. True incremental gross profit drops to $42,560. Subtract spend and the campaign cleared roughly $7,400, not $21,040.
The 4.4 ROAS was real. It was also misleading. The real lesson: cut retargeting spend by 40 percent and reinvest in prospecting where lift is highest.
ROAS in an AI ad platform
In a connected ad-creation-and-launch platform like Coinis, ROAS is the headline metric on every campaign view.
The platform does three things automatically:
- Live ROAS per creative. Spend pulls from the ad account API. Revenue pulls from your pixel and server-side events. ROAS updates per ad, ad set, and campaign without manual reporting.
- Blended vs platform-reported view. Meta's reported ROAS sits next to the blended number from your analytics stack. The gap between them is the over-attribution premium.
- Auto-shift budgets toward winners. Creatives that beat your target ROAS get more spend. Creatives that fall under it get paused. The decision rule runs hourly, not weekly.
ROAS stays the metric. The work of calculating it, comparing it, and acting on it disappears.
Related terms
Frequently asked questions
What is a good ROAS?
Most ecommerce brands target a 4.0 ROAS as a break-even point after product cost, shipping, and overhead. High-margin SaaS can run profitable at 1.5 to 2.0. Low-margin retail needs 6.0 or higher. Per Statista's 2024 ad benchmark report, median Meta ROAS sits around 2.8 across industries.
How is ROAS different from ROI?
ROAS measures revenue against ad spend only. ROI measures profit against total investment. A 4.0 ROAS can still be a losing campaign if product cost, fulfillment, and overhead eat the margin. ROI accounts for those costs. ROAS is faster to calculate. ROI is more honest about whether the campaign actually made money.
What ROAS does Google or Meta show in the dashboard?
Both platforms report attributed ROAS, meaning revenue from conversions the platform claims credit for inside its attribution window. That number is usually higher than blended ROAS (total revenue / total ad spend). Always compare both. If platform-reported ROAS is 5.0 but blended is 1.8, the platform is over-claiming.
Can ROAS be negative?
ROAS itself cannot go below zero, the lowest is 0.0 when ads produce no revenue. But the underlying campaign can lose money at any ROAS below the break-even threshold. A 1.5 ROAS sounds positive but loses money for any business with a contribution margin under 67 percent.
How does Coinis track ROAS?
Coinis pulls spend from connected ad accounts and revenue from your pixel and Conversions API events. ROAS updates in near real time per campaign, ad set, and creative. The platform also surfaces blended ROAS alongside platform-reported ROAS so the gap between the two is always visible.