What is customer LTV and how do you calculate it?

LTV — Lifetime Value or customer lifetime value — is the total economic value a customer generates for a business throughout their entire relationship with it.

But the technical definition is the easy part. What really matters is understanding what it is used for and which business decisions it changes. Because LTV is not a reporting metric — it is a decision-making tool. Without it, most decisions about acquisition, retention and marketing investment are made blindly.

In short: knowing your customers’ average LTV is only the beginning. The real value comes when you connect it with segmentation, CAC and retention actions. That is what this article is about.

Why is LTV important?

LTV answers three business questions that no other metric can answer on its own:

How much can I spend to acquire a new customer?
LTV sets the economic ceiling for CAC (Customer Acquisition Cost). Without LTV, CAC is a number without context. With LTV, it has a rational limit. An important nuance: LTV does not mean that all of that value can be allocated to acquisition — there needs to be room for margin, operating costs, structure and risk.

Where should I invest my retention efforts?
Not all customers have the same LTV. Identifying which segments generate the most value over time allows you to prioritise where to allocate resources for retention, automations and customer service.

Is my business sustainable?
The LTV/CAC ratio is one of the most useful indicators of business health. The benchmarks below are much more meaningful when LTV incorporates at least gross margin — ideally contribution margin — and when CAC and LTV are calculated on comparable bases. An LTV calculated from revenue compared against a real CAC can lead to incorrect conclusions.

An important distinction before calculating

Before discussing formulas, it is important to clarify that LTV is not a single concept — there are three versions that should not be mixed:

Historical or realised LTV — the value the customer has already generated up to today. It is the easiest to calculate because it works with real data. Its limitation is that it looks backwards.

Projected LTV — an estimate of the total value the customer is expected to generate throughout their entire relationship with the business. It combines historical data with assumptions about future purchase frequency and relationship duration.

Predicted future value — an estimate of what the customer could still spend from today onwards. This is what platforms such as Shopify calculate when they classify customers based on their future spending potential.

Mixing these three versions within the same analysis is one of the most common mistakes — and the one that leads to incorrect conclusions.

How to calculate LTV: from the basic formula to the one that really matters

Basic formula

LTV = Average order value × Annual purchase frequency × Customer lifespan

Example: an ecommerce business with an average order value of €40, a purchase frequency of 3 times per year and an average customer lifespan of 2 years.

LTV = €40 × 3 × 2 = €240

This figure represents the projected gross value. It is useful as an initial reference, but insufficient for making investment decisions.

Margin-adjusted LTV — more useful for decision-making

LTV calculated only based on sales can be misleading. A business with high revenue but low margins may have a real customer value that is much lower than the gross figure suggests.

Adjusted LTV = Average order value × Gross margin × Purchase frequency × Customer lifespan

Using the same example with a 40% gross margin:

Adjusted LTV = €40 × 0.40 × 3 × 2 = €96

Margin-adjusted LTV is more useful for CAC decisions because it is closer to the economic value a customer actually generates. For acquisition decisions, the closer we get to contribution margin — deducting variable costs associated with serving the customer, such as returns, logistics or customer service — the more useful the resulting LTV will be.

And for businesses with extensive historical data?

When there is enough data available, the most accurate LTV analysis does not come from a formula but from working with cohorts — groups of customers acquired during the same period — or with predictive models that incorporate real behaviour patterns. In ecommerce businesses with irregular purchases, seasonality or reactivation campaigns, assuming a fixed retention rate can be an excessive simplification.

The most common mistake: calculating LTV using the wrong population

The mistake is not in the formula, but in the denominator.

When calculating Customer Lifetime Value, contacts who have never made a purchase should not be included in the customer population used to calculate the average. If your database contains 20,000 contacts but only 4,000 have purchased, including the remaining 16,000 does not produce a “distorted LTV” — it produces a figure based on the wrong population that does not reflect the real value of any customer.

The correct approach always starts with customers who have made at least one purchase. And the truly useful analysis goes one step further: segmenting LTV by customer type — because the overall average hides huge differences between segments.

Average LTV is only the beginning: the value lies in segmentation

A global average LTV tells you how much an average customer is worth. But that average does not represent anyone in particular — it combines your best customers with those who have only purchased once and never returned.

The real thesis is this: average LTV matters less than understanding which customers generate value, which ones are losing value and what investment makes sense for each customer type.

LTV segmentation reveals where the real value lies, which segments require active retention, where reactivation opportunities exist and where to focus customer loyalty resources.

And there is a framework that turns this data into real strategic prioritisation: value per customer × number of customers × improvement potential. The most valuable segment per customer is not always the biggest opportunity — you need to combine all three variables to decide where to act first.

How to analyse LTV in Shopify

If your ecommerce business is on Shopify, the platform already provides native tools to analyse customer value: RFM segmentation, cohorts and predicted spending tier classification. Shopify Sidekick allows you to query this data using natural language and build reports without needing to export data to Excel. At VILAX, we work by combining Shopify’s predicted spending tier with each customer’s accumulated historical spend to identify growth opportunities within the existing customer base. This combination is what reveals the case study we describe below.

The real case: what LTV reveals when analysed by segments

At VILAX, we apply this analysis to the ecommerce businesses we manage. The data from one of our clients perfectly illustrates why the global average can lead to incorrect conclusions — and why segmentation completely changes the way we interpret the data.

The denominator problem

The average customer value from the global analysis was €14.53. A figure that, at first glance, appears low.

However, this number included more than 16,000 contacts who had never made a purchase. When calculated correctly — based on customers with at least one completed order — the average value per active customer was in a very different range, between €50 and €100 depending on the segment. The mistake was not in the metric itself, but in the denominator.

Historical spend by RFM segment

RFM segmentation — Recency, Frequency and Monetary value — classifies customers according to their actual purchasing behaviour. When applied, the analysis revealed a very clear distribution of accumulated historical spend by customer type:

The proximity between the historical spend of Champions and Previously Loyal is not an error — it is a characteristic of the RFM model. Both segments have accumulated a similar level of spend, but what radically differentiates them is recency: Champions have purchased recently, while Previously Loyal customers have not done so for some time. That is why their accumulated spend is similar, but their strategic situation is the opposite: one requires retention, the other requires reactivation.

Applying the value × volume × improvement potential framework, the At Risk segment accounted for more than €36,000 in accumulated historical spend. These are customers who already know the brand and have demonstrated spending capacity — recovering part of them represented a clear opportunity before increasing investment in acquiring new customers.

Historical spend by predicted spending tier

Shopify classifies customers according to their future spending potential into three levels — HIGH, MEDIUM and LOW. By crossing this classification with the accumulated historical spend of each group, the data revealed very significant differences:

This is where the value × volume × improvement potential framework becomes especially relevant. The MEDIUM tier has lower historical spend than HIGH — but it includes 4,081 customers compared to 1,434 in HIGH. Small improvements in purchase frequency or average order value within this segment have a much greater overall impact than any action focused on a smaller segment, even if that segment is more valuable per customer.

The opportunity was to increase the future value of MEDIUM customers through higher purchase frequency, post-purchase upselling and well-calibrated repurchase automations.

The conclusion that changes the strategy

Segment analysis revealed something that the global average would never have shown: the greatest immediate improvement opportunity was not acquiring more new customers, but retaining and reactivating the customers who already existed.

There was clear value in the At Risk and Previously Loyal segments before increasing investment in acquisition. And the priority for Champions and Loyal customers was not to sell more — it was to maintain the purchase frequency they already had.

Without segment analysis, these decisions cannot be made based on real data.

What decisions does knowing LTV change?

LTV is meaningless if it is not translated into concrete actions. These are the decisions it should drive:

Define the maximum CAC according to the customer profile. For acquisition decisions, the most useful LTV is not the RFM segment LTV — which is only known after observing behaviour — but the one calculated based on variables known from the moment of acquisition: acquisition channel, first-purchase product, category or market. RFM segments are more useful for later retention and reactivation decisions.

Identify priority reactivation segments. Segments with high historical spend that have stopped purchasing — such as At Risk and Previously Loyal — are often the most profitable opportunities. These are customers who already know the brand and have demonstrated spending capacity.

Prioritise based on value × volume × improvement potential. The most valuable segment per customer is not always the biggest opportunity. You need to combine average historical value with segment size and improvement potential to decide where to act first.

Design automations adapted to the customer lifecycle. If a customer’s value depends on purchasing every 45 days, the repurchase automation should be activated on day 40 — not day 60.

Measure the real impact of retention actions. LTV is the metric that allows you to evaluate whether a loyalty programme, a reactivation campaign or an improvement in the post-purchase experience is generating real economic returns — or just satisfaction without measurable impact.