I learned how to measure churn while working at a daily newspaper. And yet, when working with digital-first newsrooms, I often come across people looking for help with those calculations. In this guide, I’ve shared 3 different ways to measure churn for publishers across the spectrum: digital-first, newsletter writers, and even creators.
Why measuring churn is essential for publishers
Set aside any new customers you add for a second. Let’s say your churn rate is 4% per month. That doesn’t sound bad right?
This means if you start January with 1,000 paying subscribers, you’ll begin next January with 613 — a loss of 38.73%. If your average subscriber pays $60 per year, $23,220 in reader revenue is gone.
This also means the first 613 new subscribers you add during the year are only helping you break even. This is tough. Do you know what’s tougher? The underlying issues causing the 4% monthly churn rate will also affect those new subscribers.
This leads to a vicious cycle. Publishers get eager to add new subscribers quickly, so they slash prices. The problem is, discount hunters, churn at higher rates than full-price subscribers.
It’s dumping gasoline on a fire. Non-stop discounts make the churn problem worse and cut into much-needed subscriber revenue.
Accept some subscriber loss is outside your control
One of my mentors had a rule of thumb she drilled into our brains.
“Every business in every industry loses 1/3 of their customers every year.”
In my twenty-plus year career, I’ve rarely seen exceptions. For publications, this “natural” churn looks like this:
- People move to another area, so they don’t read your local news as much.
- People become parents, leading to new subjects needing their attention.
- People change careers, and they don’t need your B2B publication.
- People lose their jobs and need to cut household expenses.
- People’s interests change.
So rather than trying to have no churn, let’s accept some churn is beyond our control. In response, we focus on addressing the churn we can impact.
Every conversion point stops working someday, but it doesn’t mean you failed
Even if you aren’t churning at 4% per month or caught in the discount churn cycle, it’s still crucial to measure churn.
Everything stops working eventually. No matter how good your publication is. No matter how optimized your sign-up pages are. No matter how smoothly you convert newsletter readers to paying subscribers. At some point, they will lose effectiveness.
Granted, some conversion tactics work time and time again for years. Others may only work seasonally, or with specific audience cohorts.
But they stop working eventually. But that doesn’t mean you failed, it means the environment has changed.
This is why we need to be nimble. Keep trying new approaches to see what resonates. A diversified approach builds a strong subscriber base.
4 ways to measure churn for publishers
The Simple Method
This is the formula I use most often for the last 20 years or so. I strongly believe a metric you use is 100x superior to a metric you don’t use. If a metric is too difficult to calculate, then you aren’t going to use it.
Here’s the easiest way I know to calculate churn:
Let’s break down each piece of this formula real quick, because the definitions you choose will matter.
Defining “churned subscribers” and time period
You can define what you count as a “churned subscriber” in a way that works for you. For some publications, it may make sense to wait until after the grace period. For others, it will be more helpful to count a subscriber as churned the day they don’t renew.
However, be consistent, or you’ll be comparing apples and oranges. I usually only count the people who were active, paying subscribers on the first day of the month.
Why don’t I include people who joined and churned during the month? Because I track those separately, to make sure new subscribers are sticking around for at least a month. If they aren’t, it’s easier to act in a way that is targeted to new subscribers. There’s more detail about this in the Measuring Churn by Segment section.
You can also define the time period. I recommend sticking to monthly or quarterly. There are more useful numbers to look at week-over-week. If you wait for a year-over-year report, the root causes have gotten harder to fix. Don’t forget, churn is a lagging indicator of what’s happening.
To be consistent, I’m going to use one month as the time period in this guide.
Pros of the Simple Method
Cons of the Simple Method
For example, in July you begin the month with 1,000 subscribers
- Of those existing subscribers, 50 canceled, for a 5% churn rate (50 divided by 1,000).
- You add 400 subscribers in July, so you begin August with 1,400 subscribers.
- In August, another 50 subscribers canceled, the same as in July.
- Our churn rate for August is 50 divided by 1,400 which is 3.6%.
The lower rate makes it look like you improved! But — you lost the same number of subscribers in both months. This is why it’s helpful to report the underlying numbers with percentages.
The Averaged Month Method
One way to smooth out the effect of growth spurts is to use the Averaged Month Method.
Here’s the formula:
For the Averaged Month Method, you need to include churned subscribers who joined during the month.
Let’s break down an example, using the numbers from the Simple Method
- We start July with 1,000 subscribers.
- Of those existing subscribers, 50 canceled.
- In July, we add 400 new subscribers.
- Of those new subscribers, zero cancel (to keep the scenario the same).
This means our number of churned subscribers is 50. To get the denominator, we add the subscribers at the start of July (1,000) and the subscribers at the end of July (1,400). That’s 2,400. Divide by 2 to average out the month, and we get 1,200.
Divide 50 by 1,200 to get a 4.167% churn rate for July.
Pros of the Averaged Month Method
Cons of the Averaged Month Method
The Customer Days Method
This is the most complex formula I’ve used to calculate churn. It’s a method I stumbled across in a Shopify Engineering blog post some years ago. Rather than screwing up the formula, you can see it near the bottom of the linked Shopify post.
The gist is you think of each day as an opportunity for a subscriber to churn. So each day they don’t churn, you count those as a customer day. These customer days replace total subscribers in the period in the denominator.
That’s a simplified overview, but this method isn’t complex to set up in a spreadsheet.
This formula attempts to create a weighted average churn rate that helps predict future churn. I have two major issues with this:
- Predictions aren’t measurements.
- Churn is a lagging indicator, so “predicting” churn doesn’t work.
Besides, you need two months of data to make this calculation, so you’re still looking backward.
This is definitely a more advanced way of calculating churn. Is it more valuable than the other two methods? It depends on your capacity and goals. For most publishers, especially those under 100,000 subscribers, it’s probably overkill.
Pros of the Calculated Days Method
Cons of the Calculated Days Method
Measuring Churn by Segment
Measuring churn is a significant metric. It directly impacts reader revenue. However, to get more actionable insights, measure your churn by segment.
You’re probably already aware longer-term subscribers churn less frequently than new ones. This is one way of measuring churn by segment.
Here are helpful segments to measure churn for publishers
- Measure churn by new subscribers. I mentioned this one above. I like to keep an eye on new subscribers for the first few months. This helps make sure that the publication is doing a reliable job of connecting expectations and the product.
- Measure churn by order source. Chances are people who subscribe after reading the newsletter churn at a different rate than people who joined from a digital ad campaign.
- Measure churn by former subscribers who come back. Are people coming back to hop on the deepest discounts before quitting again? Are there seasonal patterns you can leverage for growth?
You can also apply churn to your free newsletter subscribers
Using a simple churn formula is useful for quantifying inactivity on your newsletter list.
This is especially useful when you are running pay-per-click acquisition campaigns. If people sign-up but are inactive within a month, how can I fix this? There’s probably a disconnect between the sign-up pitch and what the new subscriber expects from the newsletter.