ROAS (Return on Ad Spend) Calculator

ROAS is attributed revenue divided by ad spend. It is the standard efficiency measure for paid media, and it is also routinely misread, because a good ROAS depends entirely on your margin and a reported ROAS depends entirely on your attribution model.

How to use it

1. Enter total revenue attributed to the advertising.
2. Enter total advertising spend for the same period.
3. Read the ROAS as a multiple.

Break-even ROAS depends on your margin

There is no universal target. The break-even point is one divided by your gross margin, and everything above it is profit contribution.

At a 20 percent gross margin, break-even ROAS is 5.0 — five dollars of revenue per dollar spent just to cover cost of goods and the ad. At 40 percent it is 2.5. At 70 percent, typical of software or digital products, it is about 1.43.

That range is why the widely repeated 4x benchmark is close to meaningless in isolation. A 4x ROAS is strongly profitable for a software business and loses money for a low-margin retailer. Anyone quoting a ROAS target without stating a margin is quoting a number that cannot be evaluated.

Break-even also has to cover more than cost of goods if the campaign is meant to contribute to the business rather than merely wash its face. Payment processing, fulfilment, returns, and support all come out of gross margin before anything reaches overhead.

ROAS, ROI, and ACOS

These get used interchangeably and mean different things, which produces a lot of confused reporting.

ROAS is revenue divided by spend, so it is always greater than 1 for any campaign generating sales, and 1.0 means you recovered the ad cost and nothing more. ROI is profit divided by cost, so it is negative for an unprofitable campaign and speaks to the bottom line rather than the top. ACOS, common on retail marketplaces, is spend divided by revenue — the reciprocal of ROAS expressed as a percentage, so a 25 percent ACOS is a 4x ROAS.

Mixing them in one report is the usual source of arguments in which everyone is quoting correct figures at cross purposes. Fixing the definition in the report header solves it.

The attribution problem

ROAS is only as trustworthy as the link between an ad and a sale, and that link has become considerably weaker.

Last-click attribution credits the final touchpoint, which systematically over-credits branded search and retargeting — channels that capture demand others created — and under-credits the awareness activity that started the process. First-click has the mirror problem. Multi-touch models distribute credit but require complete cross-device data that mostly no longer exists.

Privacy changes made this materially worse. Third-party cookie restrictions, mobile tracking opt-outs, and cross-device journeys mean a substantial share of conversions cannot be tied to an impression at all. Platforms have filled the gap with modelled conversions, which are statistical estimates rather than observed events. Each platform models independently and each is incentivised to claim credit, which is why the sum of platform-reported revenue routinely exceeds total actual revenue, sometimes by a wide margin.

The practical response is to treat platform ROAS as a directional signal for optimisation within a channel, and to judge overall performance against a blended figure: total revenue divided by total marketing spend, taken from your own accounts. Blended numbers cannot tell you which channel to cut, but they cannot be double-counted either. Incrementality testing — geographic holdouts, spend pauses, controlled experiments — is the only reliable way to establish what advertising actually caused.

When a low ROAS is the right answer

Optimising for ROAS alone drives spend toward the cheapest, most easily converted audiences and away from growth. A campaign can achieve an excellent ROAS by advertising only to people who were going to buy anyway.

Two situations justify accepting a lower figure deliberately. If customers repeat, first-purchase ROAS understates the return: at a 3x lifetime repeat rate, a 1.5x first-order ROAS is a 4.5x lifetime return, and businesses that understand their lifetime value can outbid competitors who only look at the first order. And in a growth phase, buying customers at thin or negative contribution can be rational if retention is proven and the payback period is understood.

Both arguments depend on measured retention rather than assumed retention, which is where they usually go wrong.

At a glance

Frequently asked questions

Read more

Pricing and pay — Margin is not markup, a 20 percent discount can halve your profit, and a 40-hour week does not contain 40 billable hours.

Related calculators