Customer lifetime value sounds like a metric for companies with a finance team. It is more modest than that. At its simplest, it is an estimate of how much profit one customer produces over the time they buy from you, and it answers a practical question: how much can you afford to spend to win one?
You do not need software. Three numbers and a calculator will produce a figure useful enough to guide decisions.
Start with three numbers
Average order value: total revenue in a period divided by the number of orders. Purchase frequency: orders per customer per year. Retention: how long a typical customer keeps buying, expressed in years.
If your records are thin, estimate from memory and mark the figure as an estimate. A rough number you use is better than a precise one you do not have. Update it after a quarter of real records.
The simple model
Average order value multiplied by purchases per year gives annual revenue per customer. Multiply by the number of years they stay, and you have lifetime revenue.
Then apply your gross margin. If a customer spends a hundred per order, orders four times a year and stays three years, lifetime revenue is twelve hundred. At a forty percent margin, lifetime profit is four hundred and eighty. That second figure is the one that matters for spending decisions.
The arithmetic is not the difficulty. Choosing honest inputs is. Most businesses overstate frequency and understate churn, because the customers who leave are the ones nobody remembers.
Lifetime value is not a fact about your customers. It is a working assumption that improves as your records do.
Margin changes the answer
Revenue-based lifetime value flatters businesses with high costs of delivery. If two customers spend the same amount but one requires twice the support, they are not worth the same.
Where you can, split by customer type or by service. A maintenance contract and a one-off repair may share a customer but not a margin. Comparing the two tells you where to invest effort, and it often reveals that the smaller, quieter accounts are the more profitable ones.
Retention is the biggest lever
Small changes in retention move the total more than small changes in price. A customer who stays four years instead of two doubles lifetime value without any increase in order size.
So look at what makes people leave. A missed follow-up, a slow reply, a service that was never explained. Measure repeat purchases by cohort: of the customers who bought in a given month, how many bought again within six months? That single ratio is more actionable than an average.
Use it to set a sensible acquisition cost
If lifetime profit is four hundred and eighty, spending a large share of it to acquire one customer leaves little for delivery, staff and risk. A common approach is to set a ceiling well below lifetime profit, then test whether the channel can deliver customers under that ceiling.
This is where lifetime value meets paid advertising. A business that knows its figure can judge whether a click price is workable. A business that does not is guessing, and usually guesses with the revenue number rather than the profit number.
It is easier to approve a second van, a better booking system or an extra phone line when you can express the decision in lifetime terms. If a faster response keeps ten customers a year who would otherwise leave, the cost of the improvement can be compared with the profit those customers produce.
This is also the argument for spending on the unglamorous parts: onboarding, follow-up, complaint handling. They do not produce a sale, but they extend the relationship that produces several.
Mistakes that make the number useless
Using revenue instead of profit. Ignoring discounts and returns. Assuming customers stay forever because the oldest ones have. Averaging across customer types that behave nothing alike. And recalculating every month with a different method, which makes the trend meaningless.
Choose a method, write it down, and use the same one each quarter. Change the method only when you have a reason you can explain.
The lifetime value checklist
- Average order value calculated from a real period.
- Purchases per year counted, not assumed.
- Retention estimated from records, with leavers included.
- Gross margin applied, not just revenue.
- Customer types split where behaviour differs.
- Repeat purchase rate tracked by monthly cohort.
- An acquisition cost ceiling set below lifetime profit.
- Method written down and reviewed quarterly, not monthly.
The value of this model is not the figure at the end. It is the argument it starts: why do customers leave, which ones are worth more, and what would have to change for the number to double. Those questions are more useful than the spreadsheet that prompted them.
