Return on ad spend (ROAS) is one of the most talked-about marketing metrics and one of the most commonly miscalculated — usually in a direction that makes the number look better than the underlying reality. Here's how to calculate it in a way that actually tells you whether marketing is working, and what tends to move it.
The basic formula (and where it goes wrong)
ROAS is simple in principle: revenue generated from a campaign, divided by what you spent on that campaign. A $10,000 campaign that generates $40,000 in attributed revenue is a 4x ROAS.
The problem is almost never the formula — it's what counts as "revenue generated" and how it gets attributed. Two common issues:
Counting revenue that isn't actually incremental. If someone was going to find you anyway (a branded search for your exact business name, for instance) and that gets counted as ad-driven revenue, your ROAS looks better than the campaign's real contribution. This is especially common with retargeting and branded search campaigns, which convert well but often capture demand that already existed.
Using revenue instead of profit. A 4x ROAS on a service with thin margins can be far less healthy than a 2x ROAS on a high-margin service. Revenue-based ROAS is easy to calculate and useful for comparing channels to each other, but it doesn't tell you whether the business is actually more profitable — for that you need to know your margin on the jobs being booked.
A more honest way to calculate it
- Separate branded from non-branded performance, at minimum for search campaigns. Someone searching your exact business name was likely going to find you regardless of the ad; that conversion tells you less about the campaign's true value than a non-branded, generic-term conversion does.
- Track down to booked and completed jobs, not just leads or form fills. A lead isn't revenue — for ROAS to mean anything, it needs to reflect what leads actually turned into. This requires closing the loop between your ad platform and your actual sales/booking data, which is more setup work but is the difference between a real number and a vanity one.
- Where possible, calculate a profit-adjusted version too. Even a rough margin estimate applied to revenue-based ROAS gives you a much more honest picture of whether a channel is actually making the business money, not just generating top-line revenue.
What actually moves ROAS (in rough order of leverage)
Close rate, more than almost anything else. Two businesses running identical campaigns with identical cost-per-lead can have wildly different ROAS purely based on how well their team follows up and closes. Speed-to-lead (how fast you respond after someone submits a form or calls) is one of the single biggest, most under-invested levers here — response within five minutes converts dramatically better than response within an hour, and most businesses respond far slower than they think they do.
Landing page relevance and conversion rate. A page that speaks directly to the exact ad someone clicked converts meaningfully better than sending everyone to a generic homepage. This is often a bigger lever than adjusting bids or budgets.
Audience and keyword precision. Broad targeting brings in more volume at a lower average quality; tighter targeting brings in less volume but higher intent. The right balance depends on your close rate and capacity, not a universal rule.
Offer and creative. What you're actually offering (a free consultation, a specific discount, a clear guarantee) and how it's presented has a real, measurable effect on both click-through and conversion rate — and it's one of the cheapest things to test.
Budget and bid strategy, which is usually the first (and sometimes only) lever people reach for, and often has less impact than the four above once a campaign has enough data to have exited its early learning phase.
The takeaway
If your ROAS looks great but the business doesn't feel like it's growing the way the number suggests, it's worth auditing how that number is actually calculated before assuming the marketing itself is broken. And if you want to genuinely improve it, look at close rate and speed-to-lead before you touch the ad account — for most local service businesses, that's where the real ceiling is.