The CAC Formula

blended CAC      = all sales & marketing spend / new customers won
paid CAC         = paid media spend / new customers won by paid media
fully-loaded CAC = all S&M spend incl. salaries, tools, content / new customers won

The three variants differ by up to 3x in a typical early-stage business, because paid media is usually a minority of the acquisition budget. Whichever one you use, use the same one every month — a metric that changes definition between reports cannot be trended.

The rule of thumb: if the cost exists to create a new customer, it is in; if it exists to keep or serve one, it is out. Splitting a shared salary across both is fine as long as the split is documented and stable.

This is the error that makes CAC unusable rather than merely imprecise. Spend and wins are recorded in the month they happen, but the spend that produced a win happened earlier.

A company with a 90-day sales cycle doubles its ad budget in January. Wins barely move in January and February, then jump in April. Monthly CAC looks terrible for three months and excellent in April — and none of those numbers describes the business.

Two fixes, both fine:

• Trailing window. Divide the last twelve months of S&M spend by the last twelve months of new customers. The lag cancels out and the number is stable enough to trend.
• Shifted window. Compare this quarter's spend with next quarter's wins, using your measured sales cycle. Useful for a single channel, fragile for a whole business.

Whatever you pick, do not mix methods between periods. Cohort cost — spend assigned to the cohort it actually produced — is the most accurate version and the hardest to maintain.

Last quarter a company spent $180,000 on sales and marketing:

salaries + commissions    $120,000
paid media                 $35,000
content + agency           $15,000
events                     $10,000
new customers won             120

Blended CAC is $180,000 / 120 = $1,500. Paid CAC is $35,000 / 120 = $292 — arithmetically correct and strategically useless, because it pretends the other $145,000 did not exist. Fully-loaded CAC equals blended here because every line above is acquisition spend; the two diverge when a company separates brand or post-sale functions.

With ARPU of $150 a month and an 80% gross margin, monthly gross profit per customer is $120 and payback is 12.5 months. That is the number worth acting on — see CAC payback period for the thresholds, or run it in the CAC payback calculator.

• Blended is the honest company-level number and the one to use with investors. It hides which channel is doing the work.
• Paid is the right number for channel decisions, and the one that gets quoted in marketing decks because it is small. A paid CAC of $292 beside a blended CAC of $1,500 means 80% of acquisition is not paid — either organic is genuinely carrying the business, or the spend is not working.
• Fully-loaded is blended plus the overhead people forget to allocate. It is the right basis for LTV:CAC, because LTV is a gross-profit figure and the comparison should be like for like.

Report blended and fully-loaded monthly; report paid CAC per channel. Never benchmark your paid CAC against someone else's blended CAC.

• Excluding salaries. Counting ad spend only is the most common way CAC gets understated by an order of magnitude in a sales-led business.
• Counting reactivations as new. Customers who cancelled and came back cost little to re-acquire and inflate the denominator.
• Counting trials or signups. The denominator is paying customers, not leads. Lead cost is a different metric with a different benchmark.
• Putting brand spend in paid CAC. Brand campaigns do not convert in the same period and belong in blended, not paid.

There is a reverse distortion too: a one-off event or a new sales hire creates CAC spikes that are real but temporary. Track a trailing twelve months before drawing a conclusion, and compare it with the SaaS magic number, which uses the same spend but measures output as new ARR rather than customer count.

CAC on its own has no benchmark: it is only meaningful against what a customer is worth. Three checks, in order:

• LTV:CAC of 3:1 or better — the standard target. Below 1:1 you lose money on every customer; above 5:1 usually means you are underinvesting in growth. See what is a good LTV:CAC ratio.
• CAC payback under 12 months — how long gross profit takes to return the acquisition cost. Under 12 months growth is broadly self-funding; beyond 18 months every customer has to be financed from cash.
• Magic number above 0.75 — new ARR divided by the S&M spend that produced it.

Run all three in the LTV:CAC calculator, and compare your figures against segment benchmarks in the benchmark tables.