How customer lifetime value is calculated
Begin with what a single purchase is really worth: the order value less the cost of the goods and the cost of serving that order. Multiply it by how often a typical customer buys in a year, then by how long they keep buying before they drift away. What comes out is the gross value of the relationship. Take off the cost of keeping them — support, discounts, loyalty perks — and you have a number you can safely spend against.
Two inputs do most of the work: margin and retention. A business with thin margins and one-off buyers has a modest lifetime value however large the basket. A subscription, a clinic with follow-up visits, or a remittance service used every payday can carry a high one on a small first transaction. That is why an advertising cost which looks reckless for one company is a bargain for its neighbour.
Most small businesses do not need a predictive model. A backward-looking figure taken from real order history, split by acquisition channel and by product, is accurate enough to decide with and easy to defend to a sceptical owner.
Why customer lifetime value matters
It sets the ceiling on what you can afford to pay for a customer. Without it, every conversation about advertising becomes an argument over whether a lead feels expensive. With it the question is arithmetic: does the value of the relationship comfortably exceed the cost of acquiring that customer, and does it exceed it soon enough for the bank balance to cope.
It also changes which customers you chase. Channels do not deliver identical people. A cheap lead source that produces one-off discount hunters can be worth less than a dearer source that brings buyers who return. You only see the difference by following customers past their first order, which is a reporting decision as much as a marketing one.
Common mistakes with customer lifetime value
Using revenue instead of margin is the most damaging. Revenue-based lifetime value flatters every channel and encourages spending the accounts cannot support, especially where delivery, payment fees and returns eat into each order. Where cash on delivery is normal, refused parcels quietly reduce realised value further, so build the figure on orders actually paid for.
The other errors are subtler. A single blended figure hides the fact that your best segment and your worst may be nothing alike. An over-long horizon lets a spreadsheet promise value years away that the business will never bank. And a figure calculated once and never revisited stops matching reality as prices, products and churn move.
How to act on it
Work out a defensible figure per segment, agree the horizon with whoever owns the money, and write both down. Set target acquisition costs from that rather than from instinct, and allow campaigns that bring genuinely valuable customers to pay more per lead. Where the platforms accept it, feed the value back in, so bidding optimises towards customers worth having rather than towards the cheapest conversions.
Finally, treat lifetime value as something you can raise, not only measure. Better onboarding, a reason to return, a service reminder, a second product — these lift it without touching the ad budget, and they make every channel look more affordable. If the underlying numbers are not visible today, an analytics and tracking setup that ties orders back to their source is the first job.
