Stop getting high on ROAS

Return on advertising spend can help explain campaign performance, but it cannot tell you whether a business is growing profitably on its own. Revenue, margin, customer mix and the way credit is assigned all change what the number means.

Understand the calculation

ROAS divides attributed revenue by advertising spend. If a campaign spends 10,000 and is credited with 50,000 in revenue, its reported ROAS is 5x. That calculation does not deduct product costs, fulfilment, returns, agency fees or overhead.

Consider a hypothetical retailer with a 30% gross margin on that revenue. The 50,000 produces 15,000 in gross profit before the 10,000 advertising cost, leaving 5,000 before other applicable costs. The same 5x ROAS with a different margin can produce a very different result.

Define cost ratios explicitly too. Media spend divided by revenue is not interchangeable with a finance team's cost-to-income ratio covering different costs and income. Agree the numerator, denominator and accounting period before comparing them.

Ask which customers generated the revenue

A campaign can report strong returns by reaching people who already know the business or were likely to buy anyway. Those orders may be valuable, but they do not establish that the campaign created new demand.

Separate new-customer acquisition, repeat purchase and reactivation where reliable records allow it. Review customer groups over time rather than assuming every attributed order represents the same commercial contribution.

Discounts complicate this further. They can improve conversion while reducing contribution. Include the relevant discount and fulfilment economics in the assessment rather than celebrating revenue alone.

Know how the platform assigns credit

Attribution settings determine which interactions can receive credit and within what time window. Different platforms can claim the same purchase because they observe different parts of the journey and apply different rules.

That does not make every platform report useless. It makes the report a measurement view with assumptions. Reconcile it with order or CRM records, and explain where matching, consent or modelling limits the comparison. Our attribution guide covers these distinctions.

Connect acquisition cost with customer value

Customer acquisition cost and lifetime value can help assess sustainability, provided both are defined consistently. Decide which costs are included in CAC and whether LTV means revenue, gross profit or contribution over a stated period.

Do not justify present spending using an optimistic retention forecast that has not been observed. Compare mature customer groups where possible and report uncertainty where the relationship is still developing.

Use a balanced decision framework

Keep ROAS as one useful diagnostic alongside customer acquisition, contribution, retention and business constraints. Set the reporting period to suit the decision and allow for delays between enquiry, purchase and realised value.

When the question is whether an activity caused additional business, attribution alone is insufficient. A properly designed experiment or another appropriate incrementality approach can provide different evidence; it also needs enough data and suitable controls.

The aim is not to discard familiar campaign measures. It is to make sure that a good-looking dashboard is connected to customers and economics the business can actually sustain.