Stop focusing on LTV to CAC: It’s a terrible metric for subscription apps

Let me start by telling a story.

A few years ago, I had a client, a VC-backed subscription app who were gearing up for their next fundraise. And there was one set of investors that seemed particularly interested in these three questions:

1) How long do paying customers keep paying (Customer Lifetime, CLT)

2) How much do they pay you before they churn (Lifetime Value, LTV)

3) How much does it cost you to acquire those customers (CAC)

Ultimately the investors wanted to dive into the business’s unit economics. But the pitch deck focused on unpacking one metric — LTV:CAC. Whilst these are related, they are not the same thing.

And I’ll just say it: I think focusing on LTV to CAC is a terrible way of measuring the success of most consumer subscription app businesses.

But since that day, I’ve worked with dozens of consumer subscription apps and spoken to countless teams who still stress the importance of LTV to CAC.

Let’s break down each of the metrics’s components to see why LTV:CAC isn’t a great fit for sub apps, and talk about the metrics I prefer to track instead.

Why I’m not a fan of these metrics

The chaos of tracking Customer Lifetime & Lifetime Value

Starting with Customer Lifetime. How long do paying customers keep paying you?

This is a beautiful concept, but in the subscription app world, it’s sort of just that. A concept. 

It’s not actually that easy to definitively pinpoint Customer Lifetime. Firstly it can only be done retrospectively, after a user has churned. If a customer keeps paying month after month, year after year; you can’t say for certain what their lifetime is.

To add another layer of complexity, if your app is particularly seasonal, users may sign up every year between particular months then cancel their subscription and then come back the following year. So that churn we thought was churn actually wasn’t churn!

Even if your app isn’t seasonal, users may decide to pause their subscription for a whole host of reasons only to come back later. Trying out a competitor, cutting back on frivolous costs, not relating the need your app solves at that point in time. The list goes on.

All of this makes tracking Customer Lifetime incredibly difficult.

And if we don’t know what the Customer Lifetime is, how can we possibly state what our customers’ lifetime value is? We can’t.

The beauty of Customer Lifetime & Lifetime Value

Ironically, part of the beauty of consumer subscription apps is this complex and almost toxic relationship between the customer, the cancel button and the resubscribe button.

You can have users who keep paying for years on end and are incredibly valuable to your business. These customers bring up the value of any given cohort and ultimately increase the perceived lifetime value.

On the flip side you have users who cancel after just one month or even one week. These customers bring down the lifetime of any given cohort and decrease the perceived lifetime value.

So how can we monitor unit economics over time and assess the efficiency of our marketing budget without going against the natural dynamics of subscription app retention curves. 

Well, here’s what I do.

The metrics I prefer instead

Periodic ARPPU Snapshots

Rather than setting off on a wild goose chase to calculate customer LTV, I prefer to take snapshots of the revenue generated from individual customers and cohorts at specific times.

I like to call these periodic ARPPU snapshots. 

Others call this realized LTV.

Typically I’d want to take these snapshots in a way that lines up with core conversion and renewal periods. Although, these can be taken as frequently as you like.

This might look like:

• Day 0 (Trial Start / Initial Conversion)
• Day 7 (Trial Conversion)
• Day 30 (Monthly renewal
• Day 90 (Quarterly Renewal)
• Day 365 (Annual Renewal)

Alongside these core periods, I find it helpful to look at ARPPU monthly when doing more granular analyses.

In RevenueCat, this can be done really easily by filtering by Realized Customer LTV (which is the same as ARPPU) in the Cohort Explorer.

The idea here is that at certain points in time we are stepping back and saying, “Hey, right now – at this point in time – how much are customers actually worth to us?”

We understand that their value will change over time. Of course. But by taking regular snapshots, we can observe how customer value shifts over time, allowing us to take action and allocate budgets more effectively.

It also forces us to take a cohort based approach to analysis which among other things, has the added benefit of making it pretty quick to spot seasonal trends.

E.g. “Whilst it appears that Summer cohorts churn at a faster rate than Winter cohorts, they actually just follow a different resurrection pattern. They’re also much more likely to purchase consumables. So overall Day 365 ARPPU is pretty similar”

When focusing on a single LTV:CAC metric, these kinds of insights might not be so obvious.

Payback period

Once we’ve started tracking these snapshots, it also becomes much easier to dive into unit economics and identify our payback period.

Personally, I like to track ARPPU:CAC each time we take an ARPPU snapshot to measure revenue efficiency over time. But the advantages of this approach go beyond this. 

Where this really shines is for more granular cohort analysis in line with major user acquisition changes. For example, if we make significant budget shifts or increases across Paid UA, test new creative formats, launch new web to app funnels, or lean in on different JTBDs, we can take more frequent ARPPU snapshots and track the impact of our changes across blended metrics.

Gross contribution after CAC

Let’s be honest, we aren’t doing all of this work, juggling all of these metrics and reading well written articles in our free time just to break even. 🙃

We are looking to be profitable.

And that’s where Gross Contribution after CAC comes in. By continuing to take these periodic snapshots, we can see how much money we are making after CAC at various stages.

We can answer questions like:

“If we break even by month 3, how much money have we made by month 12?”

“That cohort that took 8 months to break even, how much have we made from them now?“

What’s super important to remember here is that different cohorts may have different values, different payback periods and different gross contributions after CAC. But tracking these over time, across cohorts allows us to closely monitor the health of our subscription app business.

It’s time to embrace some better metrics

So, from LTV to ARRPU Snapshots and from LTV:CAC to ARPPU:CAC.

These are subtle changes, but adopting metrics that reflect the dynamics of subscription apps makes a huge difference, especially when it comes to rapidly scaling user acquisition.