Return on ad spend is revenue divided by advertising cost. It is the headline number in most shopping accounts, and it is the number most likely to make a losing campaign look like a winning one.
How return on ad spend is calculated
Take the revenue attributed to a campaign, divide it by what the campaign cost, and you have the ratio.
Two words in that sentence carry all the risk. "Attributed" means the platform decided this sale belonged to this advert, using rules you did not write. "Revenue" means the sale price rather than what you kept.
Why the platform figure flatters you
Advertising platforms count generously, and they are not being dishonest about it. They simply count every sale they can plausibly claim.
Consequently the same sale often appears in two platforms at once. Somebody who would have bought anyway, and clicked a retargeting advert on the way, is counted as a win. Meanwhile a customer who saw the advert and bought a week later in a shop is not counted at all.
So treat the reported number as one view rather than as the truth. Comparing it against your actual revenue for the month is the fastest way to see how far apart they are.
Return on ad spend uses revenue, not profit
This is where return on ad spend does the most damage, because a ratio above the level that feels safe can still lose money.
A worked example with invented round numbers, not a benchmark. Spend ten, make forty in revenue, and the ratio is four. If the goods cost twenty five to buy, delivery costs five, and payment fees take one, you kept nine and spent ten. A ratio of four looked healthy and the campaign lost money.
The fix is to work in contribution rather than revenue: what is left after the direct costs of fulfilling the order. Every business has that number, and very few advertising reports use it.
What a good return on ad spend looks like
There is no universal figure, and any number quoted without knowing your margins is guessing.
Work out the ratio at which you break even, given your own costs. Everything above that is profit and everything below it is not, which turns a vague target into an arithmetic one. Businesses with thin margins need a much higher ratio than businesses with fat ones, and both are perfectly normal.
New customers against repeat customers
A blended figure hides the most important split in most accounts.
Advertising to people who already buy from you produces an excellent ratio and very little growth, since many of them would have returned anyway. Advertising to people who have never heard of you produces a worse ratio and all of the growth.
Consequently an account optimised purely on return on ad spend drifts towards existing customers and slowly stops acquiring new ones. Splitting the report is the cure, and it usually changes the budget conversation completely.
Return on ad spend over a customer lifetime
If customers buy more than once, judging on the first order understates every campaign.
Look at what a cohort of customers acquired in a month went on to spend over the following year. That is slower and far more honest, and it often justifies campaigns that looked marginal on day one.
When return on ad spend is the wrong measure
It suits businesses selling a product at a known price. It suits lead generation poorly, because there is no revenue at the point of the enquiry.
For services, cost per qualified enquiry and the eventual close rate do the same job more honestly. Inventing a revenue figure per lead in order to produce a ratio is worse than not having one, since everybody then trusts a number somebody made up.
Which measure fits the business is a decision that comes before the campaign runs. The wider set of measures worth choosing from is a short read, and it saves this argument later.
Using it to make decisions
Three rules keep it useful.
- Judge on contribution rather than on revenue.
- Split new customers from repeat ones before drawing conclusions.
- Compare against your own break even, not against anybody else.
Then look at the trend over a period long enough to contain enough orders to mean something, and expect the ratio to fall as you scale, because efficiency always drops as you reach further from your best audience.
Scaling without wrecking it
The ratio and the volume pull against each other, and you cannot maximise both.
Decide which you want before increasing budgets. Most businesses discover they would happily accept a lower return on ad spend for considerably more total profit, and that is a legitimate choice rather than a failure. What is not legitimate is scaling while reporting only the ratio, because it will look like performance got worse.
What to check this month
Compare the revenue your platform claims against the revenue your accounts recorded. Work out your break even ratio. Split new from returning customers.
Those three take an afternoon and they usually change what you do next more than any campaign change would have. A number that is honest is worth more than a number that is high.
Write the break even ratio somewhere everybody can see it. Once a team knows the line, arguments about whether a campaign is working turn into arithmetic instead of opinion.
Frequently asked questions
How is return on ad spend calculated?
Revenue attributed to a campaign divided by what the campaign cost. Both words carry risk: attribution follows rules you did not write, and revenue is the sale price rather than what you actually kept.
Why does the platform figure look better than reality?
Because platforms count every sale they can plausibly claim. The same order often appears in two platforms at once, and customers who would have bought anyway get counted. Compare it against your recorded revenue to see the gap.
Can a good ratio still lose money?
Easily. The ratio uses revenue, not profit. Once the cost of goods, delivery and payment fees come out, a campaign with a healthy looking ratio can be losing on every order. Work in contribution instead.
What ratio should I aim for?
The one that clears your own break even, given your own costs. Thin margins need a much higher ratio than fat ones. Any figure quoted without knowing your margins is a guess dressed as a benchmark.
Should I separate new and returning customers?
Yes, and it usually changes the budget conversation. Advertising to existing customers produces a flattering ratio and little growth. Advertising to strangers produces a worse ratio and all of the growth.
Does it work for lead generation?
Poorly, because there is no revenue at the point of enquiry. Cost per qualified enquiry and your eventual close rate do the same job honestly. Inventing a revenue figure per lead is worse than having no ratio.
Why does my ratio fall when I increase budget?
Because efficiency drops as you reach beyond your best audience. The ratio and the volume pull against each other. Most businesses would accept a lower ratio for considerably more total profit, which is a choice rather than a failure.
What should I check first?
Platform revenue against your recorded revenue, your break even ratio, and the split between new and returning customers. Those three take an afternoon and usually change your next decision more than any campaign change.