CAC vs CPL vs ROAS: What Each Metric Actually Tells You

The three definitions

CPL — cost per lead

Cost per lead is the amount spent to generate one lead, calculated by dividing total ad spend by the number of leads generated. It says nothing about whether those leads were any good.

ROAS — return on ad spend

Return on ad spend is the revenue attributed to advertising divided by the amount spent on it. It is blind to any revenue the attribution model could not see and to any cost outside the ad account.

CAC — customer acquisition cost

Customer acquisition cost is the total cost of acquisition divided by the number of new customers, including agency fees, production costs, and every lead that went nowhere. It is the only one of the three that reflects the whole funnel.

What each one cannot see

CPL cannot see quality. It improves the instant you broaden targeting, and a report showing only CPL will present that as progress.

ROAS cannot see cost outside the ad account. Agency fees, creative production, and the staff time spent working leads are all excluded, so a healthy ROAS can coexist with an unprofitable channel.

CAC cannot see speed or cause. It is a whole-funnel average that tells you the result without telling you which stage produced it — which is why it is the right number to judge a business on and the wrong number to optimize a campaign with.

How they diverge

The same month is a success on ROAS, unremarkable on CPL, and unknowable on CAC until you know what a patient is worth over time. None of the three metrics can settle that question alone.

Which to use when

Optimizing a campaign this week

Use cost per qualified lead. It moves fast enough to act on and cannot be improved by accepting worse leads.

Comparing channels

Use return on ad spend, stated alongside the attribution model used and what that model systematically under-credits.

Deciding the marketing budget

Use customer acquisition cost measured against patient lifetime value. It is the only comparison that answers whether to spend more.

The comparison that actually decides it

CAC is meaningless in isolation. A $700 acquisition cost is excellent for a patient who returns for treatments over several years and ruinous for a one-off discounted session. The question is always CAC against lifetime value, not CAC against a benchmark.

Which means a clinic that does not know what a patient is worth over time cannot evaluate its marketing, no matter how good its dashboards are. That number comes from the treatment ledger, not the ad account.

Common mistakes

• Reporting ROAS without stating the attribution model behind it
• Excluding agency fees and production costs from CAC and calling it acquisition cost
• Optimizing to CPL and treating the resulting volume as growth
• Comparing CAC to an industry benchmark instead of to your own lifetime value
• Changing how a metric is calculated mid-year and comparing across the change

Questions

In short

Use CPL to steer a campaign, ROAS to compare channels with a stated model, and CAC against lifetime value to decide the budget. Each is blind to something, so a report quoting only one is not describing performance — it is choosing which part of performance to show.