Main Metrics. LTV

Let’s find out which methods of calculating LTV exist at the moment, what you can and should use this metric for, as well as some pitfalls with using it.
Main Metrics. LTV

This article is about one of the most important financial metrics, which allows you to optimize the costs for user acquisition campaigns, to plan revenue for a long time forward, to evaluate the effectiveness of the acquisition channels, to emphasis the most financially attractive user segments and many other things.

Lifetime value, or Customer lifetime value, or CLV, is the average amount of money from one user during all his “life” in a project.

No other indicator can arguably be compared with the Lifetime value in terms of the number of ways to calculate it and its importance in assessing the financial performance of an app.

In fact, LTV is a Cumulative ARPU that has reached a certain stable value. 

In other words,  LTV is the average revenue gradually brought to an app by one user.

The LTV chart looks the same as the chart of Cumulative ARPU and it is calculated for the cohorts, and its chart and value both increase in time.

Notably, the chart grows rather quickly at the beginning and then its growth slows down and stops completely after a while.

Such trend is associated with the churn of users in the game - initially, there are a lot of people in a cohort who actively make payments, and then most of the users leave the project and revenue growth decreases by their quitting.

LTV affects revenue in the direct ratio, which is evident since the more one user brings money to a project, the higher the total revenue is.

Revenue = Lifetime value * Active users

It is worth analyzing this indicator in view of different segments when working with LTV, as all the users behave and pay differently. For example, segments can be sorted by country, traffic source, or can be action-based, e.g. by completing the tutorial. LTV of all these segments will most likely vary.

Why do we need LTV?

Evaluating the quality of traffic is one of the most frequent uses of this metric. LTV provides insight into the number of payments made by one user during their entire life in a project. Therefore, when purchasing traffic, it should be considered that the customer acquisition cost (CAC) must be less than its LTV, otherwise purchasing such traffic will only inflict losses.

Specifically, it is possible to find out when exactly a user can pay off the costs of their acquisition by calculating  Lifetime value.

It will happen when LTV becomes equal to CAC. 

If, however, if the lines cross, let’s say, after 2 years of a user's presence in the project, it is also worth evaluating if buying such traffic makes sense.

We may regard LTV as the upper limit of the cost of purchased traffic, but still, it is quite risky to buy it at a price that’s equal to the Lifetime value itself.

That’s why the recommendation of holding on to the following ratio is widely spread:


Its realization indicates the viability and success rate of a product.

The revenue forecast is one more option of using LTV, especially if you pre-estimate this indicator for different segments. In this case, if you know the audience structure and Lifetime value of each segment, it is possible to count the revenue that each of them can bring after a certain period of time.

Let’s assume that we know the LTV of the users who were acquired for money and organically, therefore we can review how traffic experiments affect the revenue: what may happen if we increase the number of organic installs by launching the app in a new country, or how the total revenue will change if we reduce paid traffic.


















$ 29,1





$44,870 (+15%)

$26,800 (-31%)

How to increase LTV?

It is hardly probable to affect Lifetime value directly since it is a complex metric. So, in order to maximize it, it's worth paying attention to the indicators that influence it the most - Lifetime, which shows how much time a user “lives” in a project, and ARPU, which shows how much he pays.

On the other hand, a user’s Lifetime in a project is affected by Retention, and ARPU is made up of the proportion of payers and their average check.

Lifetime value = Lifetime * ARPU

Lifetime can be calculated as an integral of Retention, and ARPU - as a product of the Paying share and ARPPU.

As a result, we can distinguish 5 metrics that influence LTV:

  • Lifetime;
  • ARPU;
  • Retention;
  • Paying share;
  • ARPPU.

They are exactly those levers of influence over Lifetime value. It is possible to increase LTV by raising these metrics.

How to calculate Lifetime value?

There are a great many methods of calculating LTV and that is due to plenty of factors, which we’ll discuss below.

In theory, in order to calculate LTV, we need to take a cohort of users who installed an app during a certain period of time and observe the way they will pay monthly. Hereafter, when there are no more users left in the cohort due to inevitable churn, all the revenue they brought, should be divided by the number of users from the cohort.

This method has a certain problem. It involves spending pretty much time on such calculation since users can “live” inside an app for half a year, a year, or even more. Usually, no one has so much time for calculating LTV.

Moreover, a lot of changes are likely to occur inside the app in this interval of time, and new users may behave differently from the users of the cohort we are observing. This means that their Lifetime value has all the chances to change as well.

Thus the whole complexity of the calculation is based on the fact that it takes a small amount of data that is available for the short period of time that the user spent in the app.

That’s why all the calculation methods that are used in practice are not quite accurate and have their advantages and disadvantages (by the way, devtodev uses ML for predicting LTV with 95% of accuracy). 

Nevertheless, let’s look at several approaches. 

Let's start with the formula mentioned above:

Lifetime value = Lifetime * ARPU

It’s not perfect. Firstly, Lifetime itself does not have a decisive and exact method of count and, the same as Lifetime value is calculated in advance on the basis of a small amount of data.

Secondly, whilst ARPU is also constantly changing, the formula contains only one conservative value.

Another way to calculate LTV is through the Cumulative ARPU.

If we know the Cumulative ARPU values for a certain period of time, for example, for the 1st, 7th, 14th, 30th days, then by using mathematical approximation, we can complete the curve for a longer period, the limit of which LTV will exactly be.

These are by no means all methods of calculating Lifetime value and probably they are not the simplest ones, but this is an indicator that still needs to be estimated, as it combines the most important metrics, takes into account both paying and non-paying users, and is regarded to be one of the most important financial indicators.

When we know LTV, it becomes possible to understand which customers type is of the greatest value, how soon the costs of their acquisition are paid off, how much revenue new users will bring us in six months or in a year, how this revenue will change if we increase retention, and so on. Now we can see that Lifetime Value inspires us to work on the most important financial and customer metrics and plan our business for a longer time.

Read more