Should LTV use revenue or contribution margin?
The distinction between revenue-based and contribution-based LTV matters more than any other detail in this calculation. A customer generating $2,000 of revenue at 30% contribution margin has an LTV of $600, not $2,000. Using the larger number to justify acquisition spend is how brands scale into losses.
Is LTV a measurement or a forecast?
LTV is a forecast, not a measurement. You're predicting future behavior from past cohorts, and that prediction gets less reliable the further out it extends.
What time horizon should you use for LTV?
For that reason, capping LTV at a defensible horizon (12 or 24 months) produces better decisions than modeling a theoretical lifetime. A five-year LTV justifies acquisition spend you may not survive long enough to recover.
Formula
LTV = Average Order Contribution × Purchase Frequency × Expected Customer Lifespan
For subscriptions: LTV = Monthly Contribution ÷ Monthly Churn Rate. Cap the horizon at 12–24 months for decision-making regardless of what the model projects.
Why LTV matters
LTV sets the ceiling on what you can afford to pay for a customer, which makes it the single most consequential number in an acquisition strategy. Overstating it's the most common way growth-stage companies scale themselves into trouble.
The retention leverage effect
A brand at 8% repeat purchase rate with $40 order contribution has an LTV near $43. Lift repeat rate to 22% and LTV rises to roughly $51. That means you can now outbid your former self by nearly 20% for the same customer, on identical products. This is why we build retention before scaling acquisition: it raises the ceiling on everything else.
Common mistakes
Using revenue instead of contribution
The single most damaging error in this calculation. It inflates LTV by whatever your cost structure is and produces CAC targets that guarantee losses.
Modeling an unrealistic lifespan
Five-year LTV projections from eight months of data are speculation. Cap at a horizon your data actually supports.
Averaging across dissimilar cohorts
Customers acquired through discounting behave very differently from full-price customers. A blended LTV hides that and leads to overbidding on discount-acquired traffic.
Not discounting future cash flows
Revenue three years out is worth less than revenue today, particularly for companies where capital is expensive.
Where we work on this