The short version. Revenue is not the scoreboard. Contribution margin after acquisition cost is. If you cannot state your contribution margin per order, your breakeven return on ad spend and your payback period, you are not ready to raise the ad budget.
Revenue can grow while the business loses money
I have carried the full profit and loss (P&L) for a direct-to-consumer brand. I built it to $3 million in revenue across the major online marketplaces, and I have advised more than 2,000 online businesses on growth, marketing and operations.
The failure I see most often is not a weak product. It is a decent product scaled on arithmetic nobody checked.
The pattern is simple. Ad spend goes up. Revenue goes up. Everyone celebrates. A few months later the bank balance is lower than it was, and nobody can say why.
The answer is almost always in four numbers: contribution margin per order, customer acquisition cost, return on ad spend and payback period. Get them right before you spend another dollar on ads.
Start with contribution margin per order
Contribution margin is what one order leaves you after every cost that moves with that order. Not just the product. Everything variable: goods, pick and pack, shipping, payment processing, platform fees, discounts and returns. Fixed costs such as salaries and software stay out.
Here is an illustrative order. The numbers are round on purpose.
| Line item (per order) | Amount |
|---|---|
| Average order value | $60.00 |
| Cost of goods | -$18.00 |
| Pick, pack and shipping | -$9.00 |
| Payment processing (about 3%) | -$1.80 |
| Discounts | -$3.00 |
| Returns allowance | -$4.20 |
| Contribution margin before ads | $24.00 |
That is a 40% contribution margin before advertising. The gross margin on the same order, counting only the cost of goods, is 70%. Plan ad spend on the 70% and you will overspend on every order. That gap is where the money disappears.
Returns deserve their own line. Across US retail, returns were estimated at about 13% of sales in 2024 [2]. Your rate will differ by category, so measure it per product. The $4.20 allowance above (7% of order value) assumes some returned stock can be resold. If it cannot, the line gets bigger.
Blended CAC versus channel CAC
Customer acquisition cost (CAC) is marketing spend divided by new customers. The question is which spend and which customers.
- Blended CAC: all marketing spend divided by all new customers, including those who arrived through organic search, word of mouth or posts you did not pay to promote.
- Channel CAC: spend in one channel divided by the new customers that channel claims.
Both mislead, in opposite directions. Blended CAC is flattered by free customers. Channel CAC depends on attribution, and each ad platform counts conversions its own way.
Take an illustrative month. $30,000 of spend across Meta and Google brings 1,500 new customers. Of those, 600 had no paid touch at all.
- Blended CAC: $30,000 divided by 1,500 = $20.00.
- Paid CAC: $30,000 divided by 900 = $33.33.
Against a $24 contribution margin, blended CAC says each new customer earns $4.00 on the first order. Paid CAC says each paid customer loses $9.33 on the first order. Both are true. They answer different questions.
When I grew sales through paid digital marketing, blended CAC was the number I kept under tight control. Blended is the right ceiling because it is the number the bank account sees. Channel CAC is how you find out which part of the spend is carrying the rest.
The next dollar costs more than the last one
Average CAC describes the past. Marginal CAC, the cost of the extra customers that extra spend buys, describes your next budget increase.
Illustrative: at $1,000 a day you get 50 new customers, a $20 CAC. You double to $2,000 a day and get 80. Average CAC is now $25, which still looks fine. But the extra $1,000 bought only 30 customers. That is a marginal CAC of $33.33. With a $24 contribution margin, every one of those extra customers starts $9.33 behind.
This is how revenue grows while profit shrinks. The average hides the margin you are buying at the edge.
Breakeven ROAS belongs on every dashboard
Return on ad spend (ROAS) is attributed revenue divided by ad spend. It is not profit. To judge a ROAS, compute your breakeven ROAS: one divided by your contribution margin before ads. At a 40% margin, breakeven ROAS is 1 divided by 0.40, which is 2.5.
| Reported ROAS | Revenue per $1,000 of ads | Contribution | Net after ads |
|---|---|---|---|
| 2.0 | $2,000 | $800 | -$200 |
| 2.5 | $2,500 | $1,000 | $0 |
| 3.5 | $3,500 | $1,400 | +$400 |
That is before asking whether the attributed revenue would have happened anyway. And a first-order ROAS of 2.0 can still be a good decision if those customers come back. Which brings us to payback.
Payback period is a cash question
Payback period is how long a customer’s cumulative contribution margin takes to cover what you paid to acquire them.
An illustrative cohort, with a paid CAC of $33 and a first-order contribution of $24:
- After the first order: each customer is $9.00 behind.
- 30% reorder within 90 days at the same $24 margin: +$7.20 per original customer, so $1.80 behind.
- Another 20% reorder between days 91 and 180: +$4.80, so $3.00 ahead.
Payback lands at roughly five to six months.
Because I carried forecasting, cash flow and the financing of growth along with the P&L, I treat payback as a financing question as much as a marketing one. If each month’s acquisition spend takes five months to come back, then every time you scale spend you are paying for several months of customers in advance. That gap gets funded from margin, from supplier terms or from outside capital. If none of those is lined up, a profitable cohort can still empty the bank account.
Lifetime value, measured honestly
Build lifetime value (LTV) from contribution margin, not revenue, and over a fixed window such as 12 months. “Lifetime” invites optimism.
Retention usually matters more than acquisition cost. In a study of five public companies, Gupta, Lehmann and Stuart found that a 1% improvement in customer retention raised customer and firm value by roughly 3-7%. The same improvement in margin raised it by about 1%, and in acquisition cost by only a fraction of 1% [1]. Those were large firms, not small brands, so take the direction rather than the exact multiplier. The direction is clear: fix retention and margin before you fight over a few dollars of CAC.
Mistakes I see most often
- Planning ad spend on gross margin instead of contribution margin.
- Calling revenue per customer “LTV”.
- Letting a healthy blended CAC hide a paid CAC that is climbing.
- Treating platform-reported ROAS as incremental revenue.
- Being profitable at 12 months on paper and out of cash in month three.
What I would do this week
- Build a one-page contribution margin per order for your top five products, with returns as their own line.
- Calculate blended CAC and paid-only CAC for the last 90 days.
- Compute breakeven ROAS and put it at the top of every ad report.
- Pull the customers you acquired 180 days ago and measure their payback.
- Before the next budget increase, set a marginal CAC ceiling, raise spend in one step and measure what the extra spend actually bought.
References
- S. Gupta, D. R. Lehmann, and J. A. Stuart, “Valuing customers,” Journal of Marketing Research, vol. 41, no. 1, pp. 7-18, Feb. 2004.
- Appriss Retail and Deloitte, “2024 Consumer Returns in the Retail Industry,” Dec. 2024.