How to calculate customer lifetime value

Customer lifetime value is what one customer is worth across the whole relationship, and it comes in two versions that differ by more than double. A customer spending £48 an order, 3.2 times a year, for 2.5 years generates £384 of revenue and £175.10 of contribution. The first number is the one usually quoted; the second is the only one you can actually spend, and discounting it for the years you have to wait brings it down again to £147.87. This page builds all three from records you already have.

The formula, in the only version worth using

Written out in full, lifetime value is four numbers multiplied together:

Lifetime value = average order value × orders per year × years retained × contribution margin %

Drop the last term and you get the revenue version, which is the one that circulates because it is bigger:

Revenue lifetime value = average order value × orders per year × years retained

The difference is not cosmetic. Revenue is money that passes through your account on its way to suppliers, carriers and payment processors. Contribution is what is left when those have been paid, and it is the only part available to cover advertising, overheads and you. A shop that plans against the revenue figure is budgeting money it will never see.

Use contribution rather than gross margin if you can, meaning the price less every cost that exists only because the order happened: goods, packaging, postage and the platform and payment fees. That is the same contribution figure used in customer acquisition cost, and keeping the two definitions identical is what makes the comparison between them mean anything.

Getting each input from your own records

All four inputs come out of sales data you already hold. No estimates are needed, and estimates are exactly what makes most published lifetime value figures worthless.

For a shop too young to have a year of history, the honest move is to say so and use the months you have, labelling the result as provisional. A lifetime value built on eleven weeks of data is a guess wearing a decimal point.

A worked customer

Kestrel and Fern, the invented candle maker from the acquisition cost guide, has a full year of orders. Reading the four inputs off the sales export:

Now both versions, step by step:

Revenue LTV = £48.00 × 3.2 × 2.5 = £384.00
Contribution LTV = £384.00 × 0.456 = £175.10

It is worth seeing it as an annual flow as well, because that is how it actually arrives:

The gap between £384.00 and £175.10 is £208.90 of goods, packing, postage and fees. Anyone setting an advertising budget against the £384 is committing to spend money that belongs to their suppliers.

Lifetime value against acquisition cost

The ratio is the reason the measure exists. Against the £20.00 acquisition cost worked out in the CAC guide:

LTV to CAC = £175.10 ÷ £20.00 = 8.8 to 1

On the face of it that is a very healthy business. But notice what happens if the comparison is made with the wrong numerator: £384.00 ÷ £20.00 gives 19.2 to 1, and a shop believing that figure would cheerfully double its advertising. The denominator has to be contribution, or the ratio is fiction.

The second thing the ratio hides is when. Spread the same relationship over time and set the acquisition cost against it:

The 8.8 to 1 is real, and 89% of it arrives after the first order. That is the whole cash problem of a growing shop in one line: the spending happens in month one and the return happens over thirty months, so the faster you grow the more of your own money is out in the field at any moment.

Why a pound in year three is not a pound

£175.10 treats money arriving two and a half years out as identical to money arriving today. It is not, for three separate reasons: you could have used that cash meanwhile, the customer might lapse earlier than the average says, and your own costs will have risen.

The standard correction is to discount future amounts by a rate per year. The rate is a judgement, not a fact, so pick one, state it, and be consistent.

Present value = amount ÷ (1 + rate)^years

At an illustrative 10% a year, applied to the same flow:

Discounting removes £27.23, or 15.6% of the total. The ratio against acquisition cost falls from 8.8 to 1 to 7.4 to 1 — still strong, and now honest about the waiting.

The correction matters most exactly where people lean on lifetime value hardest. A shop justifying a high acquisition cost on the grounds that customers stay for five years is counting money from 2031, and at 10% a year a pound then is worth about 62 pence now. The longer the retention assumption, the more of the answer is coming from years nobody can see.

What lifetime value cannot tell you

Lifetime value is four estimates multiplied together, which means the errors multiply too. It is a planning number and it should never be mistaken for a measurement.

• It is a forecast wearing the clothes of a fact. Three of the four inputs describe the past and are being used to predict the future. If repeat rates are drifting down, last year’s 3.2 orders is an overstatement, and the error is multiplied by the retention term.
• The average customer does not exist. Real shops have a small group who buy constantly and a long tail who buy once. A £175.10 average can be built from a handful at £900 and hundreds at £40, and an acquisition budget set on the average will overpay for everybody outside the handful. Where you can, split the calculation by how customers arrived.
• It cannot see a change you have not made yet. The figure assumes the product, the prices and the competition stay as they are for two and a half years. None of those hold still.
• Retention from a churn rate is a rough instrument. Dividing 1 by the churn rate assumes customers lapse at a steady rate. Most do not: the heaviest drop-off is right after the first order, so the average overstates how long a newly acquired customer really stays.
• It says nothing about capacity. A high lifetime value is only bankable if you can serve those repeat orders. Stock that is not there when the customer returns converts a loyal buyer into a lapsed one, which is why the measure has to be read next to when to reorder and what the stock is earning.
• It is not a licence to spend. An 8.8 to 1 ratio says the relationship is profitable over thirty months; it does not say you can afford the advertising this month. Those are different questions and only a cash forecast answers the second one.

Arithmetic and general information only — not financial, tax, legal or investment advice. Your own figures, and your own accountant, decide what any of this means for you.

Frequently asked questions

What is the formula for customer lifetime value?

Lifetime value = average order value × orders per year × years retained × contribution margin. A customer spending £48.00 an order, 3.2 times a year, for 2.5 years at a 45.6% contribution margin is worth £175.10. Leaving out the margin term gives the revenue version, £384.00, which is not money you can spend.

How do I work out how long a customer stays?

Divide 1 by the annual churn rate, where churn is the share of last year’s customers who did not buy again this year. If 40% lapse each year, average retention is 1 ÷ 0.40 = 2.5 years. Treat it as approximate: most customers lapse hardest straight after the first order, so a steady-rate average overstates how long a new customer really stays.

What is a good LTV to CAC ratio?

There is no universal figure, and the ratio is only meaningful if the top half is contribution rather than revenue. In the worked example £175.10 against a £20.00 acquisition cost is 8.8 to 1, while the same shop using its £384.00 revenue figure would claim 19.2 to 1 and over-spend accordingly. Compare like with like before comparing against anyone else.

Should I discount future lifetime value?

It is worth doing whenever the retention assumption runs beyond a year or so, because money arriving in 2029 is not money now. Dividing each year by (1 + rate) to the power of the years, at an illustrative 10%, brings £175.10 down to £147.87 — a 15.6% reduction. The rate is a judgement rather than a fact, so state which one you used.

Why is my lifetime value so much lower than figures I see quoted?

Usually because the quoted figure is revenue and yours is contribution, and the gap between the two is every cost of actually delivering the orders — £208.90 of the £384.00 in the worked example. The other common cause is a retention assumption taken from a different kind of business, since the term multiplies straight through the answer.

Other guides

• How much does Etsy take from a sale?
• How to build a cash flow forecast
• Agency utilisation rate: the formula, and the denominator that decides it