CAC payback: the one number for a Series A

What is the formula?

Take one completed period, usually a quarter.

CAC payback in months = fully loaded acquisition cost for new customers, divided by the monthly recurring revenue those customers brought, times your gross margin.

The 2019 SaaS Private Survey that KBCM Technology Group ran with forEntrepreneurs states the definition the same way: the number of months of subscription gross profit required to recover the fully loaded cost of acquiring a customer. Both qualifiers matter, and both are where the number usually goes wrong.

A worked example, with round numbers so you can follow the arithmetic:

• Fully loaded sales and marketing cost in the quarter: $600,000

• New customers won in the quarter: 40

• New monthly recurring revenue from those 40: $60,000, so $1,500 each

• Gross margin: 78 percent

CAC is $600,000 divided by 40, which is $15,000. Monthly gross profit per customer is $1,500 times 0.78, which is $1,170. Payback is $15,000 divided by $1,170, which is 12.8 months.

Two notes on practice. Match the spend to the cohort it bought, not to the calendar, if your sales cycle runs longer than a month. And compute it monthly on a rolling three-month basis, because one large deal will swing a single quarter.

What goes in the numerator?

Everything you would stop paying for if you stopped acquiring customers. The parts that get missed:

• Salaries and benefits for everyone in marketing and sales, loaded, including the fraction of a founder's time that is really sales

• Sales commission and any bonus tied to new business

• Agency and contractor fees, including creative and content production

• Software that exists to acquire customers: the ad platforms' tools, the CRM, the enrichment, the landing page builder, the analytics

• Events, sponsorships, and the travel attached to them

• The cost of acquiring customers who then churned inside the period

The last one is both the most commonly omitted and the most damaging: a funnel that wins 40 customers and loses 8 inside the quarter spent its money 48 times and is reporting 40. Customer success and support belong in gross margin rather than the numerator. Decide once, write it down, and do not move the line between periods.

Which ranges pass?

Here is what is sourced and what is judgment, kept apart on purpose.

Sourced. The guideline most founders have heard, under 12 months, comes from David Skok and was written in 2011. Skok has since said on his own site that it came from a period when capital was scarce, that land-and-expand businesses can work at around 20 months, and that sub-12-month recovery is now very rare to find. Lifetime value above 3 times CAC is his other rule, from the same page.

Benchmarkit's 2024 B2B SaaS Benchmark Report, run with Pavilion across roughly 936 companies on calendar 2023 numbers, put the median new-customer CAC ratio at $1.76: $1.76 of sales and marketing spend per $1 of new annual recurring revenue. Run that through the formula at an 80 percent gross margin and the median company needs about 26 months of gross profit to break even, or about 28 at 75 percent. That arithmetic is mine from their median rather than a figure they publish, so redo it with your own margin.

Judgment, mine. The bar I work to, and the one I have watched investors apply:

• Self-serve or SMB, where the customer can leave in a month: under 12 months, because you do not get to assume a long life.

• Mid-market at Series A: under 18 months. This is the one that gets asked about most.

• Enterprise, land and expand: 24 months is defensible if net revenue retention is comfortably above 100 percent and you can show it. Without that, 24 months is a story about the future.

• Over 30 months with retention below 100 percent is not a payback period. It is a subsidy.

The comparison worth making is against yourself a year ago and against your own best channel. If one channel pays back in 8 months and the blend is 19, you have a reallocation to make and a benchmark argument to skip.

Which three levers shorten it?

Only three things move the number. Taking the worked example above, at 12.8 months:

1. Price and billing terms. Raise price 10 percent and monthly revenue per customer becomes $1,650, gross profit $1,287, payback 11.7 months. No new traffic, no new headcount, shipped in a day. Billing terms are the other half. Annual up front does not change the metric, because the convention uses monthly gross profit, but it changes the cash entirely: an 18-month payback with annual prepayment has the money in the bank on day one. Say which of the two you mean, because investors ask about the metric and companies run on the cash.

2. Mix. Payback is an average, and averages hide a 2 to 3 times spread. Compute it per channel and per segment, then move spend. If enterprise pays back in 9 months and SMB in 22, shifting a fifth of the budget improves the blend without improving anything anywhere, which is a real gain and also a thing to say out loud in the board deck. Expansion belongs here too: in the same Benchmarkit report the median expansion CAC ratio was $1.00 against $1.76 for new customers.

3. The share of the spend that becomes a paying customer. Same budget, more customers. If trial to paid goes from 8 percent to 11 percent and the count moves from 40 to 55, CAC falls to $10,909 and payback to 9.3 months. This is the slowest of the three and the most durable, because it improves every channel at once. The method is in Trial to paid, and upstream of it in activation.

Not a lever: cutting the budget. Spending less cuts CAC and new revenue at roughly the same rate, leaves the ratio where it was, and costs you the quarter. Cutting one channel whose own payback is bad does work, and that is lever 2 wearing a different hat.

How does the number get flattered?

Two ways, and both are usually honest mistakes rather than dishonesty. Using the same example:

An incomplete numerator. Count only media spend, $260,000 of the $600,000, and CAC becomes $6,500 and payback 5.6 months. The number has halved and nothing has changed. It happens most often when the report comes out of the ad platforms, which can only see what you spent with them. Any payback number you can produce without opening the payroll file is wrong.

An inflated denominator. Two versions. Use revenue instead of gross profit and payback reads 10 months instead of 12.8. Use blended rather than new-customer acquisition, so customers who arrived through word of mouth and search sit against paid spend, and it gets worse: add 25 of them and CAC falls to $9,231 and payback to 7.9 months. In the Benchmarkit medians the same gap shows up as a blended CAC ratio of $1.61 against $1.76 for new customers.

Report new-customer payback on gross profit as the headline. Keep the blend beside it if you like, but never report it alone.