When a founder tells me their CAC, my first question is what is in it. Usually the answer is ad spend divided by new customers. That number is useful for a media buyer, but it is not customer acquisition cost, and it is not the number an investor or a CFO will calculate when they look at your books. I have already written about how to use CAC payback once you have the right inputs. This piece is about the input itself: what fully loaded CAC includes, where teams understate it, and why blended and paid CAC should always be reported together.
The definition, and why ad spend alone is not CAC
Fully loaded CAC is all the cost of acquiring new customers in a period, divided by the number of new customers acquired in that period. The word that matters is all. Andreessen Horowitz, in its widely cited list of startup metrics, says CAC should be the full cost of acquiring users stated on a per-user basis, and warns specifically against leaving out costs such as referral fees, credits and discounts. Ad spend is one line in that cost, and in a sales-assisted business it is often not the biggest line. If you employ SDRs and account executives, their salaries and commissions are acquisition cost. If you pay an agency to run campaigns, that fee is acquisition cost. If you give a first-year discount to close deals, that is acquisition cost too. A media-only number answers a narrower question: how efficiently are my ads buying customers. That is worth tracking, but it should be labelled media CAC or cost per acquisition, not CAC. Mixing the two is how a company ends up telling investors one number and discovering in diligence that the real one is several times higher. Decide the definition once, write it down, and use it consistently in every board deck.
What to include: the fully loaded CAC checklist
These are the cost lines I include, all for the same period as the customer count. 1. Paid media across every channel: search, social, LinkedIn, programmatic, marketplaces. 2. Marketing salaries, benefits and payroll costs for people working on acquisition. 3. Sales salaries, benefits and commissions for new-business roles: SDRs, AEs, sales leadership in proportion to time on new business. 4. Agency, freelancer and consultant fees for acquisition work, including creative production. 5. Marketing and sales software: CRM seats, automation, enrichment, call recording, landing page tools, attributable share of analytics tools. 6. Content, events, sponsorships and trade shows. 7. Referral fees, partner commissions and affiliate payouts. 8. Discounts, free months, credits and incentives used to win new customers. 9. Sales engineering or pre-sales time spent on new deals, if material. What I exclude: customer success, account management and expansion costs, which belong to retention and expansion economics, and general overhead like office rent, unless you have a reasonable allocation rule your finance team already uses. When someone splits time between new business and existing customers, allocate by an agreed percentage and keep it stable quarter to quarter.
Blended CAC vs paid CAC: two numbers, two questions
Blended CAC divides total acquisition cost by all new customers, including those who came organically, through referrals or by word of mouth. Paid CAC divides paid acquisition cost by only the customers attributed to paid channels. The a16z guidance is clear that blended CAC is not wrong, but it does not tell you whether your paid campaigns are working or profitable, and that investors usually treat paid CAC as more important when judging whether growth can scale. The reason is simple. A company with strong word of mouth can show a low blended CAC while every rupee or dollar of paid spend is losing money. Add more paid spend and the blended number climbs, because the organic base does not grow with the budget. Report both, every month, side by side. Then break paid CAC down by channel and track volume by channel as well, because cost per customer usually rises as you push a channel to reach a bigger audience. If you only look at blended CAC, you will not see that rise until the whole number has moved. Attribution for paid CAC is never perfect, which is why I pair it with the incrementality checks described in my piece on blended vs incremental ROAS.
Where teams understate fully loaded CAC
Four patterns come up again and again in the companies I audit. First, the founder's own selling time is excluded. In early growth-stage companies the founder often closes the biggest deals, and that cost is real even if it does not show up as a commission. I do not insist on putting a salary figure on it, but I do flag it, because CAC will jump the day a hired sales leader takes over. Second, timing mismatch. Spend in one month produces customers in a later month when the sales cycle is long, so dividing one month's cost by the same month's customers swings wildly. Use a trailing quarter, or lag the cost by your median sales cycle. Third, discounts hidden in revenue. If you give new customers a first-year discount, that is acquisition cost even though finance records it as lower revenue. Fourth, tool and agency costs sitting in general admin instead of sales and marketing. Each of these makes CAC look healthier than it is. None of them is unusual, and none is fraud. They are just definitions nobody wrote down. The fix is a one-page CAC definition, agreed with finance, that names every cost line and the allocation rule for shared roles.
How to use fully loaded CAC once you have it
A correct CAC is the input for three decisions. The first is payback: fully loaded CAC divided by gross-margin-adjusted monthly revenue per new customer tells you how many months a customer takes to repay the cost of acquiring them. I walk through that formula and a worked example in my CAC payback article, so I will not repeat it here. The second is the LTV to CAC ratio, which only means something if both sides use honest inputs; an inflated LTV over a media-only CAC is the most flattering and least useful number in a growth deck. The third is channel and budget decisions. When paid CAC by channel rises faster than blended CAC, that is the signal a channel is saturating, and the next increment of budget will cost more than the last. Use the CAC to LTV calculator and the CAC payback matrix linked below to test scenarios before changing spend. And recalculate quarterly with the same definition. A CAC number is most valuable as a trend measured consistently, far more than as a single snapshot.
Sources
Andreessen Horowitz, 16 Startup Metrics (definitions of CAC, blended vs paid CAC, and the warning about omitted costs): https://a16z.com/16-startup-metrics/
FAQ
Yes, for roles working on new-customer acquisition: marketing staff, SDRs, AEs and commissions. Allocate shared roles by an agreed percentage. Leaving salaries out turns CAC into media cost per acquisition.
Blended CAC divides total acquisition cost by all new customers, including organic ones. Paid CAC divides paid acquisition cost by customers attributed to paid channels. Paid CAC shows whether paid growth can scale profitably.
Discounts, free months and credits used to win new customers are part of the cost of acquiring them. Andreessen Horowitz lists them among the costs teams commonly forget.
Use a trailing quarter, or lag cost by your median sales cycle, so spend is matched to the customers it actually produced. Single-month CAC swings too much in sales-led businesses.