Why most CAC calculations are wrong

Customer acquisition cost is the most-cited number in growth marketing and the most-miscalculated. In the diligence work we have done across roughly forty growth programs in the last two years, three errors recur with enough frequency to call them structural rather than accidental. Each one moves the reported CAC by a margin that changes investment decisions. All three are fixable inside a quarter once the team agrees they are errors.

Error one: confusing revenue with contribution margin

The most common CAC calculation in the wild divides paid spend by new customers acquired. The number that comes out is then compared to first-order revenue to determine whether the unit economics work. This is the calculation that gets cited in board decks.

The problem is that first-order revenue is not the money the business gets to keep. It is the top of a stack of deductions. Cost of goods, payment processing, shipping, returns, gift-card breakage, promotional discounts, and the finance team's accruals all sit between the revenue line and the contribution margin. The contribution margin is the money that can actually be spent on acquiring the next customer. CAC has to be evaluated against contribution margin, not against revenue.

The size of the gap varies by category. A direct-to-consumer apparel brand with a thirty-five percent gross margin and a fifteen percent return rate has a contribution margin per first order that is roughly half of the revenue figure. A subscription software business with ninety percent gross margin and one percent churn in the first month has a contribution margin closer to the revenue figure. The same CAC dollar means very different things in those two businesses, and treating them as comparable produces nonsense conclusions.

The fix is procedural. The finance team and the growth team agree on a contribution-margin-per-first-order number, refresh it quarterly, and publish it as the denominator that CAC is evaluated against. The growth dashboards stop showing CAC versus AOV and start showing CAC versus contribution margin. The conversation in the next board meeting changes shape immediately.

Error two: ignoring the time value of repeat purchase

The second error is more subtle and more costly. Many programs compute lifetime value as a sum of future expected revenue and then compare CAC against the LTV figure to determine payback. The arithmetic is correct as far as it goes. The problem is that the LTV calculation often does not discount the future revenue against the cost of capital, and it almost never discounts it against the time the cash takes to arrive.

A subscription business with a forty-eight-dollar monthly subscription, a twenty-month average retention, and a sixty-five percent gross margin has a gross LTV of about six hundred dollars. That number gets compared to a one-hundred-fifty dollar CAC and the team concludes the business is healthy. What the calculation hides is that of the six hundred dollars of LTV, most of it arrives in months thirteen through twenty, when the customer is paying full price and not being onboarded. The present value of that revenue, discounted at the business's cost of capital, is meaningfully lower than the gross figure.

For a venture-funded business with a high cost of capital, the discount is severe. For a self-funded business with positive operating cash flow, the discount is mild. Either way, the calculation should be done. The number that matters for the question "should I spend more on acquisition" is the present-value LTV, not the gross-sum LTV.

The fix is to add a discounted cash flow layer to the LTV model. The discount rate is a conversation with finance. The rest is a spreadsheet exercise. The first time the team runs the calculation, the payback period usually lengthens, the LTV-to-CAC ratio compresses, and the case for incremental spend gets more conservative. That is the correct outcome.

Error three: blending cohorts that have nothing in common

The third error is the one that produces the most managerial heartbreak when it is caught. Growth teams routinely compute CAC as an aggregate across channels, campaigns, and cohorts. The aggregate is then used as the headline number in performance reviews. The problem is that aggregate CAC is the average of cohorts whose underlying CACs differ by factors of three or four. The aggregate is therefore not a useful management number.

A team running paid social, paid search, affiliate, and an organic-influencer program may report a blended CAC of one hundred twenty dollars. Inside that number, paid social is delivering customers at one hundred eighty dollars, paid search at sixty dollars, affiliate at ninety dollars, and the influencer program at thirty dollars. When the team is asked to cut budget by twenty percent, the answer is not "reduce blended CAC." The answer is "reallocate from the channel with the highest CAC into the channel with the highest incremental capacity at the lowest CAC, subject to the saturation curve of each."

The aggregate number is hiding the entire managerial question. Worse, when an external benchmark says "industry CAC for this category is eighty dollars," the aggregate makes the program look bad against a benchmark that itself is an aggregate of aggregates. The comparison is meaningless.

The fix is to report CAC by channel and by cohort, on a single dashboard, with the blended number relegated to a small annotation at the bottom. The conversation shifts from "what is our CAC" to "which cohorts are profitable, which are not, and which are at their saturation point." That is the conversation the team should be having, and the unblended view forces it.

A note on what counts as paid spend

There is a fourth error that is so widespread it has almost become the default. It is the practice of computing CAC using only the paid media line and excluding the production cost of the creative, the agency or contractor fees, the analytics tooling, and the salaried headcount that runs the program. The result is a CAC number that materially understates the true cost of customer acquisition.

The honest CAC includes the fully loaded program cost divided by the customers actually acquired. The honest number is usually twenty to forty percent higher than the media-only number. The honest number is what gets compared to contribution margin and discounted LTV. The understated number gets compared to industry benchmarks that were also computed dishonestly, which produces a conversation where everyone is wrong together.

Closing

None of these errors require sophisticated analytics tooling to correct. They require an agreement between the finance team and the growth team about what CAC means, a willingness to look at unblended cohort numbers without flinching, and a discipline about what gets counted as program cost. The teams that have done this work make better capital allocation decisions and have shorter, quieter board meetings. The teams that have not done it spend more time defending the numbers than improving them.