Marketing and acquisition
Customer lifetime value (LTV) calculator
Customer lifetime value is the total an average customer spends with you before they stop. Multiply average order value by how often they order in a year, then by how many years they stay. The result is revenue, not profit, which matters more than it sounds.
Net of discounts and refunds
Actual repeat rate, including one-time buyers
Softest input here. Take the shorter estimate
The formula
- Average order value
- Net revenue per order after discounts and refunds. Take it from actual orders over a reasonable window rather than from a price list, since promotional activity and mix usually pull the real figure below the one anyone quotes from memory.
- Orders per year
- How often an average customer actually buys, measured from your own order history. This is where hope tends to enter the model. Count the customers who bought once and never returned, because they are part of the average and excluding them inflates every number downstream.
- Lifespan in years
- How long a customer keeps buying before they stop. The softest input on the page and the one most likely to be wrong. If you have a churn rate, dividing 1 by it gives a rough lifespan, though that shortcut assumes churn stays flat, which it rarely does. When in doubt, take the shorter estimate. An LTV that turns out to be conservative costs you a slower quarter, and one that turns out to be optimistic funds an acquisition programme that loses money on every customer it wins.
A worked example
The outdoor gear brand has a 250 average order. Customers buy roughly 1.6 times a year and stay about two and a half years, so lifetime value comes to 1,000. Set against the 167 it costs to acquire one, that is a 6:1 ratio and the obvious conclusion is to spend more on acquisition immediately.
That 1,000 is revenue. The brand keeps 25 percent of revenue after the cost of goods, so the contribution an average customer actually delivers is 250. Against the same 167 acquisition cost the ratio is 1.5:1, not 6:1. The same two numbers, read correctly, describe a business that is barely covering the cost of growth rather than one that should be pouring money into it. Comparing a revenue LTV against a fully loaded CAC is the most common and most expensive error in this whole area, because the two figures are not measured on the same basis.
Then examine the two and a half years. If it came from a churn rate measured over one good quarter, or from the first cohort of customers who were the most enthusiastic the brand will ever have, it is probably generous. At eighteen months instead, contribution falls to 150 and the brand loses money on every customer it acquires. Nothing in the calculation would announce that. It sits entirely inside one input nobody can measure directly.
When to use it
The practical use of LTV is setting a ceiling on acquisition. It tells you what a customer is worth, which tells you what you can afford to pay for one, which is the argument behind almost every budget decision in marketing. Run it on a contribution basis and it answers that question honestly. Run it on revenue and it will authorise spending the business cannot support.
It is far more useful segmented than blended. Lifetime value by acquisition channel is often the single most revealing report a growth team can produce, because the channel with the lowest acquisition cost regularly brings the customers who churn fastest. A blended figure averages those two effects into invisibility.
It is also the number that justifies retention work. Improving repeat purchase or reducing churn moves lifetime value without touching acquisition at all, and quantifying that is usually the only way retention gets funded against a campaign that promises new customers this quarter.
Where it misleads
The output is revenue and it gets treated as value. Multiply it by gross margin to get the contribution a customer really delivers, and use that figure any time it is being compared against a cost. On a 25 percent margin the correction is a factor of four, which is more than large enough to reverse the decision it informs.
An average conceals a distribution that is rarely symmetrical. In most businesses a small share of customers accounts for a large share of lifetime revenue, which pulls the mean well above the median. Acquisition does not buy average customers, it buys whoever responds, and those are frequently the lower-value ones. Where the spread is wide, the median is the safer planning figure.
The model assumes the past predicts the future. Lifespan and repeat rate are measured on customers acquired under conditions that have since changed, through channels you may no longer use, at prices you may have moved. The further the business has travelled from the period the data came from, the less the projection is worth.
It ignores time entirely. Money arriving in year three is worth less than money arriving today, and more importantly it has to be financed until it turns up. A 1,000 lifetime value that takes three years to materialise is a very different proposition to the same figure over one year, which is why payback period belongs alongside it rather than after it.
Finally, lifespan is a guess wearing the clothes of a measurement. It arrives in the model as a clean number and leaves the calculator as a precise-looking total, and the precision is entirely borrowed. Test the result at a shorter lifespan before acting on it, and if the decision changes, the decision was never really about lifetime value.
Frequently asked
What is a good customer lifetime value?
The figure means nothing on its own, because it depends on price, purchase frequency, and industry. It becomes meaningful only against what a customer costs to acquire, where roughly 3:1 on a contribution basis is the commonly cited test of a sustainable position.
Should LTV use revenue or gross profit?
Gross profit for any decision involving cost. The revenue version is easier to calculate and is what most tools report, but comparing it against a fully loaded acquisition cost compares two figures measured on different bases and overstates the health of the business, often by several times.
How do I estimate customer lifespan?
Dividing 1 by the churn rate gives a rough figure, so 40 percent annual churn implies two and a half years. That shortcut assumes churn stays constant, and in practice it is highest early and falls for surviving customers. Cohort retention curves are more accurate where the data exists, and a conservative estimate is safer than a precise one.
What is the difference between LTV and CLV?
Nothing meaningful. Lifetime value and customer lifetime value are the same measure under two names. What varies between sources is whether the figure is revenue or contribution and whether future money is discounted, and those differences matter far more than the label.
Should future revenue be discounted for time value?
For lifespans beyond two or three years, or where money is tight, it is worth doing. Revenue arriving in year four has to be financed until it arrives. For shorter lifespans the discount is usually smaller than the error in the lifespan estimate itself, so precision there is false comfort.
Numbers are the easy part
Getting the calculation right takes two inputs. Getting an organization to agree on what counts as gain, what counts as cost, and what the number should change takes considerably longer.
That translation work is most of what I do. Westphal Solutions builds websites, tools, and reporting for nonprofits and growing teams, including the kind of measurement setup that makes a number like this one worth trusting in the first place.
Start a conversation