The CAC formula fits on a single line: divide total marketing spend by the number of new customers. Everyone knows how to divide, so most people assume they've calculated CAC — when in fact they almost always end up with the wrong number. CAC is one of the most important figures in any marketing report: it answers whether you should scale the business or change direction. Get it wrong, and every decision built on top of it rests on a false foundation.
The error isn't in the division — it's in the numerator, inside the word "total." Fill in what belongs in "total" incorrectly, and raising the budget, shutting down a channel, or increasing the price all rest on a wrong number. That's why calculating CAC correctly starts not with knowing the number, but with filling in the numerator correctly.
"Total" is not just the ad budget
The most common mistake: people take CAC to mean dividing the ad budget by the number of new customers. The ad budget is just one line in the numerator. It's the most visible line, but it isn't the only one.
The real "total" includes at least four things: the ad budget, team or agency fees, tools (CRM, analytics, automation), and content production (creative, video, design). Every dollar spent to turn a stranger into a customer belongs here. People leave these lines out not on purpose, but by oversight — because they don't show up in the ad account. As a result, CAC looks smaller than it really is, and you end up scaling a channel that's actually losing money, calling it "profitable."
A simple test: list every dollar the company spent last month to turn a stranger into a paying customer. If that list contains only the ad budget, something has been left out.
CAC ≠ CPL — where people go wrong most
The second big mistake is confusing CAC with CPL. CPL (Cost Per Lead) is the price of a single lead. CAC (Customer Acquisition Cost) is the price of a single customer. Not every lead becomes a customer, so these two numbers are never equal.
At Parking24, CPL was ~$10 and 300+ Marketing Qualified Leads came in. But in the first month, 3 deals closed. Divide the budget by 300 leads and you get one number; divide the same budget by 3 customers and you get an entirely different one. CAC is always higher than CPL, because the denominator holds not leads, but only real customers.
So "CPL is cheap" proves nothing on its own. If you have plenty of cheap leads but none of them convert into customers, CAC goes through the roof. Close out the number with closed deals, not leads.
The denominator: who is a "new customer," and over what period?
Busy with the numerator, most people forget the denominator. "The number of new customers" — for which period? The ads you run this month bring in customers next month, sometimes three months later. Divide this month's spend by this month's customers and the number comes out false — because today's customers actually came from last month's spend.
The right approach is cohort accounting: tie the spend from a given period to the exact customers that spend produced. In B2B, where the sales cycle stretches to 2–6 months, this matters all the more. Otherwise CAC looks artificially high on a growing channel and artificially low on a declining one — and you make the opposite decision.
CAC on its own tells you nothing — good or bad
Is a $7 CAC good? You can't know — not until you know the LTV, or the average revenue a customer brings. CAC is always read side by side with revenue; otherwise it's just a line of expense.
At Tuzuk AI, CAC came out at ~$7, while the first contract was $4 800. In that context, $7 is tiny. Compare: before working with us, that same client had spent $6 000 on their own and gotten 0 leads — meaning their CAC was infinite. It's not the number that speaks, it's the ratio.
In B2B, the payback period counts too: you pay CAC today, the revenue arrives months later. So look at CAC together with its payback period — otherwise you'll shut down a healthy channel too early, calling it "expensive."
CAC — the referee between two funnels
CAC's most powerful use is decision-making. Two funnels are running — which one is profitable? That question is answered with CAC, not with guesswork.
At Turon Telecom, we tested two entirely different funnels in parallel: a simple lead form at $0.40 and a full address form at $4.00. A 10× difference in lead price. But the winner was chosen by CAC, not CPL — the cheap lead isn't an automatic winner. This is stage 7 of our Go-to-Market: we calculate the cost, compare the funnels, and pick the winner by the numbers. We don't sell templates — we sell testing, and the referee of every test is always CAC.
Conclusion
The formula is one line, but the discipline lives in the numerator. Three rules for a correct CAC: count every cost (not just advertising — all of it), put only real customers in the denominator (not leads), and read CAC together with LTV and the sales cycle.
Only then does CAC turn from a simple division into a decision tool — whether to scale or change direction is decided by that number. The conclusion we've drawn from 50+ projects and $1M+ in managed budget over 2 years is simple: a miscalculated CAC means a mismanaged business.