I want to talk about something that, once explained, is completely obvious, but that most people have never thought about: Lower growth means lower churn.

For most subscription revenue businesses, growth and churn are the two metrics that combine together to determine your future. How much revenue are you adding each month in the form of new subscriptions, and how much are you losing each month in the form of churn?

*Note: The third metric that impacts this is expansion revenue which is when a current customer increases (or decreases) the amount they're paying you. That's important, but normally not as important as the other two.*

In this post, I want to explain why I think in most cases, these two metrics are inversely related to one another.

## Understanding cohort analysis

When you think about churn, you probably think of it in terms of a single number. For example, your churn might be 3% per month.

That means that every month, you lose 3% of your customers (or revenue, depending on how you count churn). If you have 100 customers this month. you should expect to lose 3 of them. So you need to add 3 new customers to break even, and any additional new customers will result in net growth for the month.

But that churn metric is often misinterpreted. It's common to think that if your churn is 3%, each customer has a 3% chance of canceling each month. In reality, that 3% number is a *blended average*. Some customers have a higher likelihood of churning, some lower, and it all averages together to be 3%.

This is where cohort analysis comes into play. Cohort analysis is when you break your customers into different cohorts (or "groups" or "segments" or whatever) and compare the numbers of each cohort separately over time. Here's an example of a cohort report from Google Analytics showing what percentage of website visitors come back each day after their initial visit:

When talking about churn, you'd normally base the cohorts on signup date, where everyone who signed up in a given month is in a cohort together. Then you can see how each cohort churns each month.

## Who has the highest churn?

If you were to do that analysis, I can pretty much promise that what you'd see is very high churn the first few months after a cohort's signup date, and then it would gradually decrease over time.

Why is this? It's because the people who initially sign up aren't all equally likely to become long-term customers. Some of them are just tire-kickers who are still evaluating you. Some are really interested, but haven't actually activated (i.e. gotten fully set up). Some are 100% committed, and they're getting tremendous value from your product already.

Generally speaking, the tire kickers will churn right away. They were never really true customers to begin with. The unactivated people will likely also churn before too long (unless you can get them activated). That leaves you with just the perfect customers left over at the end. Because those people are getting so much value from your product, they're much less likely to churn.

## How this impacts the blended average

So at any given time, you have a new cohort of people who just signed up. That cohort will have a high churn rate because there are so many tire kickers and unactivated users. There will also be older cohorts who will have lower churn because many of the bad-fit users have already left. As the cohorts get older and older, the churn drops significantly.

If your churn rate is 3%, that means that the average of all of those cohorts is 3%, but the older cohorts are almost certainly well below that number. If someone has been with you for five years, their odds of canceling next month are probably close to zero.

## Why this means low growth = low churn

You've probably pieced this together already, but here's the key point: If your growth rate slows down, a lower percentage of your customers will be in those newer cohorts. Each cohort's churn will probably be about normal, but the blended average will be weighted more heavily towards the older cohorts which will pull your overall churn rate down.

Of course, the reverse is also true. If you have a big surge in growth, you should expect that because a higher percentage of your customers are new, you'll have more churn until things equalize. I've personally experienced this before where a temporary surge in signups is followed by a few mediocre months.

## What you should do with this information

I don't know! I just think this is interesting, sort of obvious (once you've thought about it) and rarely discussed. So I wanted to discuss it. Ok, mission accomplished, have a good day.