a business dies in the unit, not the plan; CAC vs LTV on one customer tells you if growth scales
Unit economics decide because arithmetic wins: if a single customer costs more to acquire and serve than they return in profit, every new customer the company adds makes it worse off. No amount of volume can turn a negative per-customer contribution into a positive aggregate one.
The mechanism is simple. Customer acquisition cost (CAC) is the total cost of turning a lead into a paying customer, including sales time, marketing spend, onboarding, and any discount or free trial. Lifetime value (LTV) is the net present value of the gross profit that customer contributes over the whole relationship, not just their first purchase. Subtract the first from the second, and you have the per-customer contribution. Multiply by a thousand customers and you have a thousand times the problem if that number is negative.
Consider a subscription service charging $10 per month. Assume the average subscriber stays 20 months, which implies a churn rate around 5% per month, and that the service carries an 80% gross margin. LTV is $10 × 20 × 0.8 = $160. If CAC runs $50, the ratio is 3.2, comfortably above the common 3x target. But if a marketing change pushes CAC to $60, the ratio drops to 2.7. A startup adding 10,000 customers at that CAC will have spent $100,000 more in acquisition than those customers will ever generate in gross profit. That cash is gone before the company knows it.
The most common misunderstanding is equating LTV with revenue. LTV must exclude the variable costs of serving the customer, otherwise it overstates health. Another error is ignoring the payback period. An LTV/CAC of 5x looks great, but if the full CAC is only recouped after four years, the business needs enough cash to fund a very long ramp. The 3x rule is a heuristic, not a law; it exists because it gives some cushion for the costs that are easy to miss, such as customer support, payment processing, and infrastructure.
Unit economics also do less work when the business is a two-sided marketplace or a platform. A marketplace may lose money on one side to attract the other, and the value of a customer depends on the presence of other customers. In that case, a simple per-customer LTV/CAC comparison understates the total picture because the customer makes the product more valuable to everyone else. Similarly, if a company sells to a large enterprise with a long sales cycle, the acquisition costs may be spread over months and the LTV may be highly variable; the unit still matters, but the ratio must be computed over cohorts, not as a single average.
None of that changes the core lesson: prove the unit works before scaling. A plan says what you hope will happen; unit economics say what is already happening for one customer. When those two conflict, believe the arithmetic.
Cram So if I get a million customers, I am rich right?
Rep Not if each one costs you more to acquire than they pay you.
Cram But with scale, costs drop...
Rep Acquisition costs might. Unit economics are per-customer math.
Cram You mean like, profit per customer?
Rep LTV minus CAC. Lifetime Value minus Customer Acquisition Cost.
Cram So LTV has to be bigger than CAC?
Rep At least 3x. Otherwise every new customer makes the problem worse.
Cram That sounds backward. Growth is supposed to be good.
Rep Growth multiplies losses if the unit bleeds. A business dies in the unit.
Cram So I should check one customer before scaling?
Rep Exactly. Prove the unit works. Then pour fuel on it.
Cram What about the plan, the vision?
Rep The plan is a story. The unit is arithmetic. Arithmetic wins.