What ROAS measures
ROAS answers one question: for each dollar you put into ads, how many dollars of revenue came back? It is the most common way to judge a paid campaign, an ad set or a single ad, because it ties spend directly to money rather than to clicks or impressions.
ROAS is a revenue ratio, not a profit ratio. A campaign with a ROAS of 3 can still lose money if your product costs, payment fees and refunds eat more than two thirds of the revenue. That is why the useful question is never "is my ROAS high?" but "is my ROAS above my break-even ROAS?"
Teams write the same number three ways. A ROAS of 4 is also 4x, 4:1 or 400%. They all mean $4 of revenue per $1 of spend.
ROAS formula
ROAS = revenue attributed to ads ÷ ad spend
- Revenue attributed to ads: the sales your attribution rules credit to the campaign in the period. Which sales count depends on the attribution model and attribution window, so two tools can report different ROAS for the same campaign.
- Ad spend: what the ad network charged for the same period. Some teams add agency fees and creative costs; if you do, say so, because it lowers the number.
To express it as a percentage, multiply by 100. A ratio of 2.5 is a ROAS of 250%.
Worked example
You spend $2,000 on a Meta campaign in March. Customers whose first visit came from that campaign pay you $7,000 in March.
ROAS = $7,000 ÷ $2,000 = 3.5 (3.5x, or 350%)
Is that good? Say your gross margin after payment fees and refunds is 60%. Your break-even ROAS is 1 ÷ 0.60 ≈ 1.67. At 3.5 the campaign returns $7,000 × 0.60 = $4,200 of gross profit for $2,000 of spend, so it earns $2,200 after paying for itself.
Now compare two campaigns that look alike in the ad platform:
| Campaign | Spend | Revenue | ROAS | Profit at 60% margin |
|---|---|---|---|---|
| Retargeting | $500 | $2,500 | 5.0 | $1,000 |
| Cold audience | $1,500 | $4,500 | 3.0 | $1,200 |
Retargeting has the higher ROAS, but the cold campaign makes more money in total and brings customers who weren't already on their way. ROAS ranks efficiency; it doesn't tell you where the next dollar does the most. Look at the profit column as well before you move budget.
What is a good ROAS?
There is no universal good ROAS, because it depends on your margins. Any ROAS above your break-even point makes money on the first sale; anything below it loses money unless customers come back and pay again.
- Physical products with thin margins need a high ROAS. At a 25% margin, break-even is 4.0.
- Software and digital products usually have high gross margins, so break-even often sits between 1.1 and 1.5. A first-month ROAS below 1 can still be fine for a subscription, if customers stay long enough.
- Subscriptions should be judged on more than the first payment. A $30/month plan that keeps customers for 12 months is worth about $360 per customer, so judge the campaign on customer lifetime value, or watch the same cohort's ROAS climb as renewals arrive.
Use the break-even ROAS calculator to find your floor, then set a target above it that leaves room for overhead.
How to measure ROAS
- Tag every ad with UTM parameters so each visit carries its campaign, ad set and ad. Ad networks also add a click ID such as
gclidorfbclid, which tells you the network but not the campaign. - Record revenue where it happens: in your payment provider (Stripe, Paddle, Shopify and so on), not in the ad platform's pixel. Refunds should come off.
- Connect each payment to a visit. Pass the visitor ID into checkout, or identify the user when they sign up, so the payment can be traced to the click that brought them.
- Pull spend from the ad account for the same dates, in the same currency, and divide.
For a quick estimate without any setup, put a campaign's revenue and spend into the ROAS calculator.
Why ad platforms report a higher ROAS
Meta Ads and Google Ads each count conversions with their own tracking and their own rules. Meta, for example, can credit a purchase made within days of a click and within a day of someone merely seeing an ad. Google Ads now uses data-driven attribution or last click. Every network that touched a customer claims the sale, so the platform-reported ROAS of your channels, added up, is usually more revenue than you actually collected.
A ROAS computed from your own payments, with each sale credited once, is lower and more honest. It is the number to set budgets with; the platform number is useful for the platform's own bidding.
Common ROAS mistakes
- Treating ROAS as profit. A ROAS of 2 on a product with a 40% margin loses money. Compare against break-even, not against 1.
- Mixing periods. Spend from the last 7 days divided by revenue from the last 30 days gives a meaningless number. Use the same dates for both.
- Counting the same sale twice. Adding up Meta-reported and Google-reported revenue double counts customers who clicked both.
- Ignoring refunds. Gross revenue flatters campaigns that attract buyers who return the product.
- Judging subscriptions on day one. The first payment is a fraction of what a subscriber is worth. Look at ROAS again after renewals come in.
- Reading ROAS on tiny numbers. One $900 sale on $100 of spend is a ROAS of 9 and tells you almost nothing yet.
How MetricRaven calculates ROAS
MetricRaven computes ROAS from your own payments, not from an ad pixel. Connect a payment provider and a Meta Ads or Google Ads account, and Revenue → Paid acquisition shows spend, revenue, ROAS and cost per customer for every campaign, ad group (Meta: ad set) and ad.
- Revenue is what customers paid in the period, net of refunds and converted to your reporting currency, credited to each customer's first tracked visit (first-touch attribution). Each sale counts once, so channels never double count.
- ROAS is that revenue divided by the network's spend for the same days, converted at the European Central Bank rate of each day.
- Network-reported ROAS (Meta-reported or Google-reported) sits next to it for comparison and is never added into revenue.
- Assisted revenue shows payments from customers who clicked an ad but first arrived some other way. It isn't part of ROAS, so you can see what a campaign helped close without double counting.
- Visits with only a click ID or the network's UTM tags that match no synced campaign appear on their own line and stay out of ROAS.
The same ROAS figure appears as a tile on the Traffic and Revenue tabs. For networks MetricRaven doesn't import spend from, filter the Paid channel by campaign and divide its revenue by the spend in your ad account. See Meta Ads, Google Ads and the Revenue dashboard docs.
Questions
- How do you calculate ROAS?
- Divide the revenue a campaign brought by what you spent on it in the same period. $6,000 of revenue from $1,500 of spend is a ROAS of 4 (4x or 400%).
- What is a good ROAS?
- Any ROAS above your break-even ROAS, which is 1 divided by your gross margin. With a 50% margin you break even at 2; with an 80% margin, at 1.25. Subscriptions can accept a lower first-payment ROAS if customers renew.
- Is ROAS the same as ROI?
- No. ROAS divides revenue by ad spend. ROI divides profit by the full investment, so it accounts for product costs and other expenses. A campaign can have a ROAS of 3 and a negative ROI.
- Why is my ROAS in Meta Ads higher than in my analytics?
- Meta credits purchases made within its attribution setting after a click or an ad view, and models some conversions it can't observe. Other networks claim the same sales. Analytics that credit each payment once, from your own payment records, report less.
- What does a ROAS of 0.8 mean?
- Each $1 of spend brought back $0.80 of revenue, so the campaign doesn't even recover its cost on the first sale. It can still work for a subscription if customers renew enough times; check its revenue again after a few billing cycles.