Two metrics that sound alike and mean very different things. Mixing them up is how "profitable" campaigns quietly lose money — the dashboard shows green while the bank balance shrinks.
Both numbers measure return, but they draw the boundary in different places. One looks only at the advertising channel; the other looks at the whole business. Knowing which is which — and which to trust when — is the single most useful piece of maths a media buyer carries. This guide defines each, puts them side by side, and shows you the trap that catches nearly everyone at least once.
ROAS is revenue divided by ad spend. A ROAS of 3× means every $1 of ads returned $3 of revenue. Its strength is speed: it uses data the ad platform reports in near real time, so it is the fastest signal of whether a campaign is heading the right way, and it is what you watch hour to hour while a test runs. Its weakness is that it measures the ad channel only — it ignores every other cost. Product cost, platform and processing fees, refunds, tools and your own time are all invisible to it. A high ROAS tells you the ads are pulling their weight; it does not tell you the business made money.
ROI is profit divided by total cost, and it counts all the costs ROAS leaves out: ad spend, tools, fees, refunds, chargebacks and your time. It is slower and messier to calculate because some of those costs only land days or weeks later, but it is the number that tells you whether the business actually made money. Where ROAS answers "is the ad channel efficient?", ROI answers "did I come out ahead after everything?" The two can point in opposite directions on the very same campaign, which is exactly why beginners get burned. Which payout model you run also shapes the ROI picture — a lifetime revenue share can turn a thin day-one number positive over time, as covered in CPA vs RevShare vs Hybrid.
Same campaign, two lenses. Read this next to analytics for beginners so the inputs behind each figure are clean before you trust either.
| Metric | Formula | What it counts | Speed | Answers |
|---|---|---|---|---|
| ROAS | Revenue ÷ ad spend | Ad spend only | Fast, near real-time | Is the ad channel pulling its weight? |
| ROI | Profit ÷ total cost | Every cost, including your time | Slower, needs full costs in | Did I actually make money? |
A 3× ROAS looks great — until you add product cost, fees and refunds and discover your ROI is negative. This is the classic failure: scaling a campaign because the channel metric is glowing, while the real return has been underwater the whole time. It is one of the most expensive beginner mistakes, and it happens because ROAS is the number the platform pushes in your face while ROI is the number you have to assemble yourself. The rule is simple: never let a healthy ROAS alone justify a budget increase.
The bridge between the two metrics is your break-even ROAS — the ROAS at which ROI is exactly zero. If your non-ad costs eat half of every dollar of revenue, you need a ROAS of roughly 2× just to break even, and anything below that loses money no matter how good it looks. Thin-margin offers demand a high break-even ROAS; fat-margin offers can be profitable at a ROAS that would bankrupt a thinner one. This is why a "good" ROAS is meaningless in the abstract: the same 3× is wildly profitable on one offer and a slow bleed on another. Work out your break-even first, then you know what a winning ROAS actually looks like for your numbers.
Use both, at different tempos. Optimise to ROAS for speed — it is the fast, in-platform signal you use to steer creatives and cut losers while a test is live, because ROI data is simply too slow to react to in the moment. But judge the business on ROI — that is the number that decides whether a campaign is worth scaling, keeping or killing. In practice you set a target ROAS that already bakes in your break-even and margin, run the account against that target day to day, and then reconcile against true ROI on a slower cadence to make sure the target is still honest. Feed both into the wider allocation decisions a buyer makes, described in what media buyers actually do.
Not on its own. A high ROAS on a thin-margin offer can still lose money once product cost, fees and refunds are counted, while a lower ROAS on a fat-margin offer can be very profitable. ROAS is only meaningful against your break-even ROAS, which depends on your margin.
Because ROI data arrives too late to steer a live campaign — refunds, chargebacks and delayed costs land days later. You optimise to ROAS for the fast in-platform signal, then reconcile to ROI on a slower cadence to confirm the business actually made money.
The ROAS at which your ROI is exactly zero. If non-ad costs consume half of every revenue dollar, you break even around 2× ROAS. Calculate it first, and it converts the vague question "is this ROAS good?" into a concrete target you can run the account against.
Watch ROAS to run the campaign day to day, but never scale on it alone. Set a target ROAS that already includes your margin and break-even, and check true ROI before pouring in budget — that is what stops a flattering channel metric from hiding a losing business.
What media buyers actually do day to day: testing creatives, reading data, managing spend and scaling the campaigns that turn a profit.
Core · 11 min readHow to structure paid campaigns: the campaign / ad set / ad hierarchy, ABO vs CBO, one concept per ad set, naming conventions, and how structure fe...
Advanced · 14 min readThe algorithm handles targeting now; creative is what you control.
