LTV is one of those concepts that can easily be understood at surface level but often misunderstood at its core. But what exactly makes it so difficult to grasp? 1. LTV is hard to calculate, and nobody knows what to do with it. 2. There’s no way easy way to use it to measure growth. 3. Fixing (1) and (2) is much easier said than done.
.png)
LTV is one of those concepts that can easily be understood at surface level but often misunderstood at its core. But what exactly makes it so difficult to grasp?
For those that don’t know, LTV stands for (Customer) Lifetime Value - the most common way to measure it being the average revenue that a user brings in (ARPU) divided by the probability that the user is going to churn.
The most straightforward way to explain this is though an example. Let’s say you have a meditation app that charges $10 per month and has an average churn rate of 25%. That means, mathematically, a typical user will pay you for four months before churning in the fifth. If you magically reduced your churn rate to 10%, then suddenly you could expect the average customer to pay you for 10 months and then churn on the 11th.
.png)
You can see the appeal of a number like this. Most customers do not pay for subscriptions forever, and it’s useful to have some way of measuring the continuous value that your subscribers will provide to your business.
There is a large issue with this methodology, however: It’s not completely straightforward to measure churn at a granular level (ARPU itself isn’t always straightforward either). You can take a cohorted retention rate, but at that point, it’s just two of your metrics divided by each other. Amazon’s online retail business has an ARPU of (1) the cost of the prime subscription, plus (2) the average order volume of the customer. It’s churn can be calculated based on a lack of orders, a cancellation of the prime membership, or a mixture of both.
The easiest LTV calculation doesn’t actually provide that much value when you think about it. All you’re really doing with the basic calculation is taking two metrics and blending them into one. Because the equation is bimodal, explanations in its movement always have to have an inclusion about whether the numerator or denominator changed and by how much. At best, it’s a way to provide an extra metric to shareholders/investors, and at worst, it’s yet another KPI to confuse your team with.
We’ve also found that lots of folks are confused on how to calculate it. Some folks keep it as a running number, while others treat it like a progress bar - they calculate the cohorted retention rate at some point of maturation in the customer journey, and then try to figure out what percentage of LTV has been captured over the course of the year.
We think that both are wrong. The first method is useless, and the second one is too confusing to explain to stakeholders. Your LTV calculation should take both value captured as well as the LTV still to capture. However, we think that the LTV “left to capture” should be taken as a probability based on current data (Markov) as opposed to something else.
Your LTV should be broken up into two parts: pLTV and fLTV.
This is LTV that’s been captured already, and is a summation of all of the money that the customer has paid for you thus far. If a customer has been a subscriber for 10 months and pays you $10/month, then your pLTV is $100.
fLTV is the LTV yet to be captured. Because this is a probability calculation, we believe that this should be recalculated on a regular basis. We also think that it should be calculated at the individual level, rather than at an aggregate. Calculating LTV this way unlocks a huge number of benefits, which we’ll go over in the next section.
Now for the fun part: How can we make the LTV useful? There are a variety of ways, from product development and UX enhancement, as well as triggers for different marketing campaigns. We’ll go over 3 below:
One of the primary viewpoints of a company’s public stock value is effectively a bet on its future earnings - and LTV is a strong predictor for future earnings (it’s in the name!). Keeping track of aggregate movements in LTV allows you to better understand which cohorts of users are poised to provide value for the company in the future.
Take a look at this calculation, based off of a completely fictional company called “Spitify” - an audio-based subscription platform that allows a user to stream sounds of different celebrities spitting. They offer three tiers: an “Individual” membership offered at $9.99/mo, a “Twice the Fun” membership tier for two users at $12.99/mo, and a “House” membership offered at $16.99/mo. Take a look at the LTV growth numbers from Spitify’s investor day this year (they track LTV changes using Cotera):
You can see there that even though total LTV is highest for the individual tier, the grouped memberships had a much larger LTV growth. This helps provide context to subscriber growth and retention benefits as well - because both of those play a factor in calculating LTV, there’s a possibility that the growth in the grouped subscription tier could come more from an increase in engagement (and subsequent reduction of churn) instead of purely subscriber growth.
You can use Cotera or a product analytics tool to figure out what usage patterns and acquisition funnels lead to higher LTV:
The graph above comes from the Cotera instance of “NedFlix” - a streaming service for exclusively showing programming starring actors named Ned. It shows that customers who have high LTV are more likely to watch more episodes on average. This may be self-explanatory, but the next finding is a little more interesting: customers who are primarily “desktop” users are more likely to fall into the “Low LTV” category when compared to mobile users. Knowing this, a product team can start to make decisions to push users more towards using their mobile device.
You can also use Cotera, or a data-warehouse integrated CDP, to push customers that match the criterion above into a marketing platform, like Braze or Attentive. For instance, lifecycle-focused marketing teams can use data on what characteristics promote high-LTV users to tailor email campaigns to folks who stand to gain a lot of value from taking those actions. In this case, a user who has hit a certain customer journey metric and has never used the Nedflix mobile app should be part of a marketing campaign that drives them to download and consume content on their mobile device.
.png)
What you shouldn’t do is assume you can apply the same marketing strategies to your entire sea of customers. This is outrageously inefficient, especially when the LTV data needed for you to make important distinctions with is within arm’s reach. Don’t get it confused either — both present and future LTV are equally as useful and should all be considered when making big decisions about your company. And most importantly, don’t be afraid to ask for help. After all, that’s what Cotera’s here for.