Key takeaways
- Churn is the share of customers who leave in a set time. To cut it, you need to know why they leave, in their own words.
- Ask at the moment people leave with a one-question exit survey, and read every answer.
- Watch for early warning signs in feedback, like "cancel," "switching to," "too expensive" or "stopped working."
- Sort the reasons into ones you can fix (bugs, gaps, confusion) and ones you can't (they no longer need it).
- Fix the top reason you can fix, tell the people who raised it, and check if churn for that reason drops.
Every app and SaaS loses customers. Some leave because their needs changed. But some leave for reasons you could have fixed, if you had known in time. Feedback is how you find out.
This guide shows you how to ask people why they leave, how to spot churn before it happens, and how to turn what you learn into fixes.
What is churn, and how do you measure it?
Churn is the share of customers who stop paying or stop using your product in a set time, usually a month. The basic formula is:
Churn rate = customers lost during the month ÷ customers at the start of the month × 100
Here's a made-up example. You start the month with 200 paying customers. During the month, 10 cancel. Your churn rate is 10 ÷ 200 × 100 = 5% for that month.
Keep two things in mind. First, count only customers who were there at the start. New sign-ups during the month don't count yet. Second, track churn by plan and by how long people have been customers. Churn in the first month can look very different from churn after a year.
See the churn rate entry in our glossary for more detail.
Why does feedback help reduce churn?
Numbers tell you how many people left. Feedback tells you why. Without the why, you are guessing.
Feedback also warns you early. Some people complain before they cancel. They email support, leave a low rating, or mention a rival in a review. If you catch those signals, you get a chance to help before they go.
How do you ask people why they're leaving?
Use a short exit survey at the moment people cancel. One question is enough. Make it multiple choice, with an optional text box.
A simple exit survey
What's the main reason you're cancelling?
- It's too expensive for me
- I'm missing a feature I need
- It was too hard to use
- Something didn't work right
- I switched to another app
- I don't need it anymore
- Something else
Anything else you'd like us to know? (optional)
Keep the list short and in plain words. Always include "something else." Read the text answers. They often hold the most useful detail.
App Store subscriptions
If you sell subscriptions through the App Store, people cancel in their Apple account settings, outside your app. So you can't show a survey at that exact moment. App Store Server Notifications can tell your server when someone turns off auto-renew. Then you can follow up by email or with an in-app message, if you have a way to reach them.
What are the early warning signs of churn in feedback?
Look for words that signal someone is thinking of leaving. Tag them, so they stand out from other feedback.
| Signal | Example words | What to do |
|---|---|---|
| Cancel intent | "cancel," "refund," "unsubscribe" | Reply fast, ideally within a day |
| Rival talk | "switching to," "X does this better" | Ask what the rival does well |
| Price pain | "too expensive," "not worth it" | Check if they use the paid features |
| Broken trust | "lost my data," "stopped working" | Treat as urgent, fix, then follow up |
| Fading use | "I don't use it much anymore" | Ask what changed, and show what's new |
Also watch the numbers next to the words. A drop in logins, a lower survey score, or a 1-star review from a paying customer are all worth a look.
Synthiya tags each piece of feedback by type, including cancellation risk. On plans with Slack alerts, it can ping you when feedback looks urgent. See plans.
How do you turn churn reasons into fixes?
Sort every reason into two groups: ones you can fix, and ones you can't.
Reasons you can fix include bugs, missing features, confusing screens and slow support. These are your to-do list.
Reasons you can't fix include "my project ended" or "I changed jobs." You can't stop these. But you can make it easy to come back later.
Then rank the fixable reasons. Count how often each one comes up, and how much revenue it costs. Pick the top one and fix it well, rather than five at once. Our guide to analyzing customer feedback shows a simple way to score themes.
Tip
Price is often a stand-in for value. "Too expensive" can mean "I didn't get enough out of it." Check how much the person used the app before you cut the price.
How do you win back people who left?
Tell them when you fix the thing that made them leave. This is the simplest win-back there is, and it feels personal because it is.
- Keep a list of who left and why.
- When you ship the fix, email the people who named that reason. Keep it short and honest.
- Make it easy to return. One link back to their account or the App Store page.
If you sell App Store subscriptions, Apple also supports win-back offers for people whose subscription has ended. Check Apple's StoreKit docs for the current rules.
A weekly churn routine
You don't need a big process. Thirty minutes a week is enough to start:
- Read every exit survey answer from the past week.
- Tag each one with a reason.
- Check your churn-risk feedback: support emails, low ratings and reviews.
- Reply to anyone who sounds ready to leave.
- Update the count for each reason.
- Pick or confirm the one reason you're fixing now.
Once a month, look at churn by reason. Did the reason you fixed shrink? If so, move on to the next one. That's the feedback loop at work.
Want a quick read on how your customers feel overall? Our guide to NPS vs CSAT explains which survey to use and when.
Frequently asked questions
What is a good churn rate?
It depends on your price, your market and who your customers are. The most useful benchmark is your own past churn. Aim to bring it down month by month.
How do I ask App Store subscribers why they cancelled?
People cancel App Store subscriptions in their Apple account settings, not in your app. App Store Server Notifications can tell your server when someone turns off auto-renew. Then you can ask by email or with an in-app message, if you have a way to reach them.
Should I offer a discount to people who want to cancel?
Sometimes. A discount can help when price is the real reason. It won't help if the app doesn't do what they need. Read the exit survey first.
How many exit survey answers do I need?
Start reading from the first answer. You may see patterns after a few dozen. Keep collecting, because reasons change as your product changes.
What is the difference between churn and retention?
They are two sides of one coin. Retention is the share of customers who stay. Churn is the share who leave.
Terms in this guide
Written by the Synthiya team. Last updated . See something wrong or out of date? Tell us and we'll fix it.
Claude by Anthropic
ChatGPT
Insight Timer: Meditate, Sleep