Understanding Churn: The Number You're Probably Measuring Wrong
Let's talk about the scariest number in your community dashboard.
Churn.
You see it tick up by a percentage point and your stomach drops. You stare at it. You try to convince yourself it's a blip. You find yourself wondering if everyone is about to leave and you just haven't noticed yet.
I want to walk you off that ledge today. Not by telling you churn doesn't matter, it absolutely does, but by helping you understand what you're actually looking at, what a healthy number looks like, and how to tell the difference between a churn rate that's genuinely bad news and one that's just a normal part of running a community.
By the end of this post, churn will still matter to you. But it won't scare you the way it does right now.
What Churn Actually Is
In the simplest terms, churn is the percentage of members who leave your community in a given period.
If you start the month with 200 members and 10 leave, your monthly churn is 5%.
Simple enough, except there are a few different ways to measure it, and choosing the wrong one will make you panic for no reason.
Gross churn. The raw percentage of members who left, regardless of whether new ones joined. This is what most dashboards show by default.
Net churn. Members who left minus new members who joined, as a percentage of where you started. If 10 left and 15 joined, your net "churn" is actually negative, i.e., you grew.
Voluntary vs involuntary churn. Voluntary churn is someone actively canceling. Involuntary churn is their payment failing and them getting kicked out by the system. These are very different problems with very different solutions.
Most creators look at gross churn and panic, when they should be looking at involuntary and voluntary separately, and watching net movement over a rolling window.
What Healthy Churn Looks Like
Here's the part nobody tells you clearly enough.
Healthy monthly churn for a consumer community is usually between 3% and 8%.
Yes, 8% churn every month can be healthy.
Most people reading that sentence have to read it twice. They assume churn should be like 1% or maybe zero if they're doing a good job. That assumption is wrong, and it comes from confusing consumer communities with B2B SaaS.
Consumer subscriptions, and most paid communities are consumer subscriptions, have structurally higher churn than enterprise software. That's because individuals' lives change. They get busy, they change jobs, they have babies, they go through a rough patch, their priorities shift. Some percentage of members will always leave for reasons that have nothing to do with you.
For context:
Netflix churn is about 2.5% monthly.
Spotify premium churn is around 4%.
Most gym memberships churn at 5-7% monthly.
Typical paid community churn runs 4-10% monthly.
If your community churn is 5%, you are not losing. You are running a normal consumer membership.
When Churn Is Actually a Problem
Now the signal inside the noise.
Churn is a problem when one of these three things is true:
1. It's accelerating. 4% churn is fine. 4% churn this month, 5% last month, and 6% the month before is a trend. Levels don't matter as much as direction. A stable 7% is less alarming than a rising 4%.
2. Early-tenure churn is high. If members are leaving in their first 30 days, you have an onboarding or expectation-setting problem. If members are leaving after a year or two, that's usually normal life turnover. Always slice your churn by tenure.
3. The reasons are concentrated. If everyone who leaves cites the same reason ("the weekly call moved to a time that doesn't work," "I wasn't finding the content useful anymore"), you have a specific problem to fix. If reasons are scattered (moved, too busy, finances, just not using it), it's the normal noise of life.
None of these are about the absolute percentage. They're about direction, cohort, and cause.
The Metric That Beats Churn: Net Revenue Retention
Here's the grown-up version of churn that I wish more creators measured.
Net Revenue Retention (NRR) asks: of the revenue you had from a cohort of members 12 months ago, how much of it do you still have today, accounting for everything (cancellations, upgrades, downgrades, new purchases from those members)?
Healthy community NRR for paid communities tends to run between 70% and 95%. Meaning, of every $1000 in revenue you had from a cohort a year ago, you still have $700 to $950 today. The drop comes from churn. The partial recovery comes from members upgrading, renewing annuals, or buying add-ons.
NRR above 100% is the holy grail, and usually only achievable with strong upgrade paths or annual plan encouragement.
If you only look at one number, look at NRR, not monthly churn. It tells the real story of your community's financial health over time.
The Emotional Part
Let me say the uncomfortable thing. Churn hits every creator emotionally in ways the dashboard never captures.
Every cancelation feels personal. You know most of your members by name. When one of them leaves, part of your brain whispers "they didn't find it valuable enough." It's an easy trap to fall into, and it's especially brutal when you're building something you care about deeply.
Here are the reframes I'd offer:
You are not the only reason people leave. People's lives are huge and complicated. Your community is one of 30 subscriptions they have. Sometimes they need to cut five of them. You didn't fail.
A member's last month doesn't define their whole tenure. Someone who stayed for 14 months and canceled because they got busy still got real value from you. Don't let the exit event rewrite the story.
Exit surveys are gold. Most creators skip them. When someone cancels, ask one question in a warm, low-pressure way: "Anything you'd want me to know about why you're leaving?" You'll get genuinely useful feedback from about 40% of people, and zero hostility.
Winbacks are real. Members who leave sometimes come back months later when their life settles down. Don't assume a cancellation is forever. Keep the door open.
What to Do If Your Churn Is Too High
If your churn is genuinely high (say, above 10% monthly or accelerating), here's a rough order of operations.
First, segment. Break churn down by tenure (0-30 days, 30-90 days, 90+ days). Most high churn is driven by one cohort. Find which.
Second, read your exit surveys. Look for concentrated patterns. One reason showing up 30% of the time is a problem to fix. No pattern means the problem is more diffuse.
Third, look at engagement in the 30 days before cancelation. Members who cancel almost always go quiet first. That quiet window is your earliest warning sign and your intervention opportunity. A warm, personal outreach in week two of silence ("hey, haven't seen you in the community lately, everything okay?") saves more subscriptions than anyone would guess.
Fourth, fix payment failures before content. A huge chunk of what creators call "churn" is actually involuntary churn, members whose cards expired or payments failed. Better dunning, retries, and pre-expiration reminders can recover 20-30% of that quietly.
Fifth, remember that some members are just not a fit. A small percentage of churn is a feature, not a bug. You don't want to retain people who don't love your community. Let them leave with a warm goodbye.
One Last Reframe
The creators with the healthiest businesses don't have zero churn. They have predictable churn that they've priced in, offset by a steady flow of new members who are a great fit.
A community with 5% monthly churn and 7% monthly new member growth is thriving. That's what "sustainable" looks like in practice.
So the next time your dashboard shows 5% churn and you feel the stomach-drop, breathe. Check the trend. Slice by tenure. Read the exit surveys. Take what's useful. Let the rest go.
Your community is probably healthier than you think. And now you'll be able to tell.