Customer Acquisition Cost

Blended, paid and fully-loaded

These are not competing attempts at the same number. They answer three different questions, and a company that reports well typically tracks all three.

• Blended CAC — all sales and marketing spend divided by all new customers. Answers: what does growth currently cost us in total? Good for board reporting, useless for channel decisions.
• Paid CAC — paid spend divided by paid-attributed customers. Answers: what does the next marginal customer cost? This is the number that should govern budget decisions.
• Fully-loaded CAC — everything, including salaries and commissions. Answers: is this acquisition motion viable at all? For a sales-led business this can be several times the paid-media figure, and it is the only version that belongs anywhere near an LTV:CAC ratio.

What belongs in the numerator

Fully loaded means fully loaded: advertising and media, agency and contractor fees, sales and marketing salaries with employer costs, sales commissions and SDR bonuses, content and design production, events and sponsorships, and the tooling that exists to support acquisition — CRM, enrichment, outbound sequencers, ad platforms.

Two categories are argued about every time. Customer success that works exclusively on renewals and existing accounts is a retention cost, not an acquisition cost; it belongs in cost of revenue or in operating expenses, and putting it in CAC quietly makes your acquisition look worse and your margin look better. Brand spend with no measurable attribution still belongs in blended CAC — it is real money spent to acquire customers — even though it will never appear in paid CAC.

What belongs in the denominator

New logos, and only new logos. Three things routinely contaminate the count. Reactivated customers are cheap to win back and counting them as new flatters CAC. Upgrades and plan changes are expansion, not acquisition. And in a product with a trial, deciding whether a customer is acquired at trial start or at first payment changes both the count and the timing — pick one, write it down, and never mix the two in a trend.

The lag problem, which is the one that bites

Spend in a given month produces customers in that month and the two after it. Divide this month's spend by this month's new customers and any period where budget is ramping shows a CAC spike that reflects timing rather than efficiency — and any period where you cut spend shows a phantom improvement.

Two fixes. Use a quarter rather than a month, which absorbs most of the lag. Or, better, match spend to the cohort it produced: attribute spend to the month it was committed, then count the customers it eventually converted. The second is more work and considerably more honest.

Why blended CAC hides the trend

Blended CAC falls whenever organic acquisition grows, even if every paid channel is getting more expensive at the same time. A company whose word-of-mouth is compounding can watch blended CAC improve for four straight quarters while the marginal cost of a paid customer nearly doubles underneath it. Track paid and blended side by side, and segment both by channel and by contract value — an average CAC across a self-serve plan and an enterprise motion describes neither.