Everybody can tell you how to calculate ROAS. Revenue divided by spend. A 4:1 is good, they add, and then the article ends. I have sat in a lot of meetings where that formula was the only thing anyone agreed on, and the meeting still ran an hour over, because three people had three different ROAS figures for the same campaign and all three were calculated correctly.
That is the part worth writing about. The arithmetic takes one line. Everything that decides whether the answer means anything sits either side of it.
What is the formula for ROAS?
Return on ad spend is the revenue attributed to your advertising divided by what you spent on it over the same period. Nine hundred euro of tracked revenue from three hundred euro of spend is a 3.0, or 300 percent, or 3:1. Three ways of writing one number, so pick a convention and hold it.
Note the two words doing all the work: attributed and tracked. Neither means banked. The formula is a ratio between a figure the platform calculated and a figure your bank calculated, and the first one is an estimate.
Why does the same campaign show three different ROAS figures?
Because three different systems are answering three slightly different questions. The ad platform reports revenue it believes its advertising caused. Your analytics reports revenue it can attribute to a session it saw. Your accounts report money that arrived. All three are measuring real things, and they will not match.
The gaps have ordinary causes. A platform counts a sale against the day the advert was seen, not the day the order was placed. Analytics loses anyone who declined tracking or arrived on a different device. The books include refunds, failed payments and the order somebody placed over the phone after seeing the advert. None of that is a bug, and chasing the three figures into agreement is a project with no end. Pick one as your decision number, write down which, and let the other two be diagnostics. Reconciling an attributed figure against money you actually received is the subject of how to measure marketing ROI.
Which revenue figure should you use?
Use the platform figure to compare one advert against another, and the banked figure to decide whether advertising is working at all. They are tools for different decisions and substituting one for the other is where most reporting arguments start.
Inside a platform, everything is measured consistently, so the comparison between two ad sets is fair even if the absolute number is generous. Across the business, only the money that arrived counts. The practical answer is to run both: optimise campaigns on platform ROAS, judge the channel on a monthly reconciliation against revenue you actually received.
What is a break-even ROAS, and how do you find yours?
Your break-even ROAS is one divided by your gross margin. If sixty cents of every euro of revenue survives the cost of delivering it, your gross margin is 0.6 and you need a ROAS of about 1.67 before advertising stops losing money. Below that figure, growth costs you money on every sale.
This single calculation is missing from almost every ROAS conversation I have been in, and it is the one that makes the number mean something. A 3.0 is excellent at a 20 percent margin and roughly break-even at a 6 percent one. Anybody telling you a 4:1 is good without asking what you sell is guessing.
How do you calculate ROAS in Meta Ads?
Meta divides the conversion value its pixel or Conversions API recorded by the spend in the same window, and reports it as purchase ROAS. The figure depends entirely on the attribution setting above the table, which you can change after the fact and which will rewrite history when you do.
So check the setting before quoting the number. A seven-day click and one-day view window will show a materially higher ROAS than a one-day click window on the same campaign, because the first credits Meta with more of the sales. Neither is dishonest. They are different definitions, and reporting a figure without the window attached is what makes the number unquotable. Where Meta ROAS typically sits, and how to read it, is in Meta ads ROAS benchmarks.
How do you calculate ROAS in Google Ads?
Google reports it as conversion value divided by cost, usually expressed as a percentage, so a 400 percent target ROAS means the same thing as a 4:1. It depends on you having assigned a value to the conversion, which is the step most accounts get wrong.
Two faults recur. Values are set once as an average order value and never updated, so the ROAS slowly detaches from reality. Or every conversion action is counted, including newsletter signups and phone clicks, which inflates the figure with things that never produced revenue. Fix the conversion values and the ROAS becomes useful without touching a bid.
Is ROAS the same as ROI?
No, and the difference is the cost of everything that is not the advert. ROAS measures revenue against media spend only. ROI measures profit against total investment, which includes the cost of what you sold, the agency or in-house time, the tools and the platform fees.
That is why a campaign can run at a healthy ROAS and still lose money. ROAS is a media efficiency metric, nothing more, and it is the right one for deciding where the next hundred euro of budget goes. For whether the activity is profitable, the fuller calculation is in how to measure marketing ROI, and for how long you can afford to wait to get the money back, allowable CAC and payback.
What are the four questions to ask before you trust a ROAS?
Ask what revenue figure it came from, what attribution window produced it, what period it covers, and what the gross margin behind it is. A ROAS quoted without those four is a number without a unit, and it cannot be compared with anything, including the same campaign last month.
- Which revenue? Platform-attributed, analytics-attributed or banked. They will differ, often by a lot.
- Which window? One-day click reads very differently from seven-day click and one-day view, and the setting is changeable.
- Which period? A fortnight of a long-cycle product measures almost nothing. Match the window to how long people take to buy.
- Which margin? Without it there is no break-even to compare against, so good and bad are opinions.
Run those four questions over your own reporting and you will usually find the disagreement was never about the arithmetic. It was that two people were using the same word for different things.
How often should you look at ROAS?
Weekly to spot something breaking, monthly to make decisions, and never daily for anything other than checking that spend is still going out. Daily ROAS mostly measures noise, and reacting to it is how accounts get edited into the ground.
The reason is sample size. Most small accounts do not generate enough conversions in a day for the figure to be stable, so yesterday’s bad ROAS is usually tomorrow’s good one with no intervention at all. Decide your review rhythm in advance and hold it, because the discipline is worth more than the sensitivity.
If you want this read over your own account rather than in the abstract, that is the kind of work my agency does.
Frequently Asked Questions
What is a good ROAS?
There is no universal figure, because a good ROAS is anything comfortably above your break-even, which is one divided by your gross margin. At a 60 percent margin that break-even is about 1.67, so a 3.0 is strong. At a 15 percent margin the break-even is nearer 6.7 and the same 3.0 loses money on every sale.
How do you calculate break-even ROAS?
Divide one by your gross margin expressed as a decimal. A 40 percent gross margin gives a break-even ROAS of 2.5, meaning you need two euro fifty of revenue for every euro of advertising before the activity stops costing you money. Use gross margin after the cost of goods and delivery, not your net profit margin.
Why is my Meta ROAS higher than my actual revenue suggests?
Usually the attribution window. A wide window credits Meta with sales that other channels or your existing audience contributed to, and view-through conversions add purchases from people who saw but never clicked. Refunds, failed payments and duplicate pixel events widen the gap further. Reconcile monthly against banked revenue and treat the platform figure as a comparison tool.
Should ROAS include VAT or tax?
Use revenue excluding tax, because tax is not yours to keep and including it inflates the figure. The rule that matters more is consistency: the conversion values you send to the platform and the revenue in your reconciliation must sit on the same basis, or the two will never agree.
Is a 4:1 ROAS really the benchmark?
No. It is a rule of thumb from retail businesses with roughly 25 percent margins, repeated until it sounded universal. It is a reasonable target if your economics happen to resemble that, and actively misleading otherwise. Calculate your own break-even first, then set a target with enough headroom above it to cover the costs ROAS ignores.



