什么是客户流失率？客户留存终极指南

客户流失率是在特定时间段内停止使用产品或服务的客户或用户的百分比，是了解业务健康状况和客户留存策略有效性的关键指标。

Analytics and Measurement

The churn rate measures how many people stop using your product or service in a given amount of time. Put simply, it's how fast your customers are leaving your company. For subscription-based businesses, this number is quite literally your life or death.

If you're shedding 5% of your customer base each month, you'll lose about 46% of your customers over a year. Which is why knowing and improving your churn rate isn't just important, it's a matter of survival in today's dog-eat-dog digital world.

The figures attributed to ProfitWell's 2024 report could not be checked because the page is no longer accessible: the often-quoted benchmarks that B2B SaaS companies churn about 4.2% a month and consumer subscription services as much as 7-10% appear mainly on third-party blogs. The value of retention, by contrast, is well documented: research by Frederick Reichheld of Bain & Company shows that increasing customer retention rates by 5% increases profits by 25% to 95% (Harvard Business Review, 2014).

Churn rate explained from A to Z

Churn rate indicates how quickly customers are leaving. Think of a leaky bucket: if the holes (churn) are too large, you will never fill it, no matter how much water (new customers) you pour in.

At its core, the basic formula looks deceptively simple:

Churn Rate = (Customers Lost During Period ÷ Customers at Start of Period) × 100

But here's where it gets interesting. Sometimes when even seasoned marketers try to compare monthly versus annual churn they get confused.

The comparison between customer churn and revenue churn is next-level, though. You might lose 30 small customers but keep one big one, which results in high customer churn but low revenue churn. Both are important, but for distinct reasons.

Then there is the distinction between gross churn and net churn. Gross churn counts only lost revenue, while net churn also takes into account expansion revenue from existing customers. A company can even achieve negative net churn, a sign that it is adding more revenue from existing customers than it is losing.

Important industry benchmarks

Acceptable churn rates differ by industry, business model and customer type. The latest research in the field from Recurly (data updated July 2026) puts the average churn rate across industries at 3.27% (2.34% voluntary and 1.25% involuntary), with SaaS at 3.22%, ecommerce at 4.25% and education at 4.99%. The other commonly cited benchmarks below could not be found there:

The often-quoted average monthly churn rate for B2B SaaS companies, 4.2% (down from 4.4% in 2023), could not be traced to an original source. Customer size matters a great deal:

Small business customers: often quoted at 3-7% monthly churn
Mid-market customers: often quoted at around 1-3% monthly churn
Enterprise customers: often quoted at under 1% monthly churn

Streaming services have higher churn. Netflix has one of the lowest monthly churn rates in the business, about 2.3-2.4% according to Argoid; the claim that Paramount+ exceeds 7% could not be traced to a source.

Subscription commerce varies widely (the figures below are often quoted but could not be traced to a source):

Health and fitness: often quoted at 10% monthly churn
Beauty and personal care: often quoted at 8%
Food and drink: often quoted at 12% monthly churn

The telecom industry is often said to have annual churn of 15% to 25%, a figure that could not be traced to a source. T-Mobile reported a record-low postpaid churn of 0.77% in the second quarter of 2023, better than both AT&T's and Verizon's for the first time, according to The Motley Fool; the linked article does not mention AI retention strategies.

Banking and financial services are often said to have annual retention rates of around 75%, meaning about a quarter of customers leave each year; this figure could not be traced to a source.

How to calculate churn rate correctly

The big problem is that the churn calculation that seems so simple is full of pitfalls that can ruin your analysis. Now onto the correct ways to do it:

Basic Monthly Churn Rate

Here's how you can work it out: (Customers lost in month ÷ Customers at start of month) × 100

If you have 1,000 customers on the first day of January and 50 of them go away, then your churn rate is 5%.

Converting to Annual Churn

Never multiply monthly churn by 12. It is the most common and serious mistake in churn analysis.

The correct formula is: 1 - (1 - monthly churn rate)^12

For instance, if 2 percent churn per month, 21.5 percent churn annually, not 24 percent.

How to Calculate Revenue Churn

You can do so with: (Revenue lost from churned customers ÷ Total revenue at the beginning of the period) × 100

Sometimes that's an entirely different story from lost customers, especially if a high-value customer leaves.

Churn Analysis by Cohort

To observe churn trends over time, segment customers by sign-up date or by characteristics. The often-quoted claim that most SaaS businesses get their churn rate down from 10% in the first month to 4% in the third month could not be found on the linked Chargebee page.

Modeling Predictive Churn

Today, companies use machine learning to spot behavior patterns that signal when people are about to leave. Some important features are:

How often you log in
Number of support tickets
How often people use features
History of payments

Real-world success stories

HubSpot's Customer Happiness Index

HubSpot was one of the first companies to measure churn risk with a score, its Customer Happiness Index (CHI), which tracked which parts of the product each customer was using, according to investor David Skok; Skok also notes that CHI was not a great predictor of churn. The page linked as HubSpot Customer Happiness Index (HubSpot's guide to churn rate) does not describe CHI.

The often-quoted claims that CHI saved 33% of unhappy customers and could identify at-risk customers 3 to 6 months before standard metrics could not be traced to a source.

The Recommendation Engine on Netflix

Netflix invests heavily in personalization: its recommender system influences choice for about 80% of hours streamed, and Netflix's own researchers estimated that personalization and recommendations save the company more than $1 billion a year (Gomez-Uribe and Hunt, 2015). Argoid puts Netflix's monthly churn at about 2.3-2.4%; the claim that the industry average for streaming is 5.8% could not be found there.

Its ability to win back customers is particularly impressive. According to Antenna data reported by Xataka, other platforms recover only 34% of cancelled subscribers on average, while Netflix recovers 50% after six months, and 61% of its cancelled subscribers return within a year.

The Value Stacking of Amazon Prime

What's unique about Amazon Prime is that it takes a value-stacking rather than a personalized approach. By combining shipping, streaming, music and reading benefits, they've created an ecosystem too important in your life to give up.

According to Consumer Intelligence Research Partners (CIRP), 94% of US Prime members renew after one year and 98% after two years (2021 data). The psychology is simple: even if a member uses only one or two benefits heavily, the whole package feels worth more than the annual fee.

T-Mobile's Record-Low Churn

Telecom operators commonly use machine learning to study customer behavior, predict churn and run personalized retention campaigns before customers leave. T-Mobile's result shows what low churn looks like: a record-low 0.77% postpaid churn in the second quarter of 2023, better than AT&T's and Verizon's for the first time, according to The Motley Fool, which does not attribute the result to AI.

Why you should care about the churn rate more than you do

Looking at how churn impacts the fundamentals of business explains why it should be the top priority. Churn Rate and CLV are negatively related. If you cut your churning in half, then your CLV is doubled.

Think about the unit economics. If your cost to acquire a customer (CAC) is $500 and each customer pays $50 a month, it takes 10 months to recoup that cost. With 5% monthly churn, the average customer stays about 20 months; at 2.5% monthly churn, about 40 months, so the same CAC generates far more profit.

The impact on valuation can be large, although the often-quoted claim that SaaS companies with sub-3% annual churn fetch 2 to 3 times the multiples of those with 10% or more churn could not be traced to a source.

i

"...it's critical to view churn for what it is: a symptom of a deeper, underlying disease. And that disease is a failure to ensure your customers achieve their Desired Outcome..."

Lincoln Murphy, founder of Sixteen Ventures, in "Churn is a Symptom, Not a Disease" (June 2016)

Companies develop a different mindset when they shift from trying to stop churn to ensuring customer success. The often-quoted claim that public SaaS companies retaining over 120% of net revenue are worth far more than their peers could not be found on the linked Lighter Capital page, although net revenue retention is widely watched by investors.

What not to do when doing math

1. Combining Different Types of Customers

When you lump trial users in with paying customers, the metrics become meaningless. Trial users usually churn at far higher rates than paying users, so mixing them distorts the numbers.

2. Not Knowing How Long A Contract Is

Customers who fail to renew their annual contracts don't leave in 11 months, but they do leave in month 12. Monthly-paying customers churn gradually in the course of a year.

3. Blindness to Seasonality

Many companies review budgets in the fourth quarter, which can raise B2B SaaS churn, and consumer subscriptions often see cancellations after the holidays. These seasonal patterns are often said to be predictable, but the page linked as Studies in the field does not discuss seasonality.

4. Treating All Churn the Same

There are two kinds of churn: voluntary churn, which occurs when clients decide to leave, and involuntary churn, which is the failure of payments. Each needs a different solution. Recurly puts average involuntary churn at 1.25% across industries, and much of it can be recovered with better payment handling, such as card updates and payment reminders.

5. Adding New Customers to the Bottom

Your churn rate is 5% (5/100) when you begin with 100 customers, add 20 and lose 5 of them. It's not 4.17% (5/120).

Proven strategies that work in reducing churn

1. Onboarding at a World-Class Level

Great onboarding is one of the most effective ways to reduce churn. The often-quoted claim that MobileAction reduced time to value by 32% and the number of new features touched by 38% with guided onboarding flows could not be found on the linked SuperOffice page.

Some important parts are:

Gradual revealing of features
Clear milestones
Personalized setup assistance
Early rewards in the first session

2. Scoring the Health of Customers

Today, most scoring models rely on one predictive score, which blends usage, lifespan, engagement, support interactions, and payment history. Automated workflows initiate interventions when health scores drop below threshold level.

3. Pricing Strategies

Longer commitments tend to reduce churn: SuperOffice cites SaaS Capital's finding that companies with longer average contract terms have 30-50% lower churn rates. The answer? Keep monthly payment options while offering meaningful discounts for paying the whole year up front.

4. Communicating Value Ahead of Time

Customers churn more often than not because they forget why they signed up. If you email them regular value realization reports that demonstrate the actual ROI, time saved, or goals met, then they'll remember why your service matters at all.

5. Building a community

Customers who take part in active user communities are often said to stay 2-3 times longer, a figure that could not be traced to a source. Community members are not just buying a product; they are joining a club.

Why you should know how to calculate churn rate

1. Higher Lifetime Value of Customers

Lower churn means higher CLV. The two move in opposite directions: halving churn doubles CLV.

2. Reduced cost of acquiring new customer

The expense of acquiring new customers goes down when they stick around. Businesses that don't shed customers frequently can afford to pay more to acquire new ones and still be profitable in the long run.

3. Consistent Revenue Growth

For one, a low churn rate can make revenue streams more predictable, which is useful for planning and investing decisions. Investors love predictability, and the more stable your retention, the more they'll pay for your company.

4. Better Unit Economics

It is an obvious point, but unit economics improve when customers remain longer. Margins rise as a result of fixed costs being divided over more revenue.

5. Improved customer feedback loop

You have customers who stick around and give you feedback all the time, and that makes you better all the time. They're the perfect partners to help you develop your products.

6. Better Position in the Market

Businesses that don't lose customers very often can then invest more money into new ideas and bringing their markets to even greater scale, providing them with durable competitive edges.

Questions that come up a lot

How much churn is "good"?

Baremetrics notes that 5-7% monthly churn is generally considered healthy, and that mature B2B SaaS companies aim for 1-2% monthly or 5-10% annual churn. Business-to-consumer (B2C) subscription services generally see higher churn. Context matters a lot, depending on the industries you are serving and the nature of your business.

How do you calculate churn for yearlong contracts?

Yearly contracts are a different story. Do churn at renewal point, not over the course of months. Monitor both customer count and revenue churn, they don't always align with yearly contracts.

What is the distinction between gross and net churn?

Gross churn is the revenue lost directly to cancellations and downgrades. Net churn offsets that loss with expansion revenue from existing customers. Negative net churn means you are earning more from existing customers than you are losing.

Should you consider free users when doing your churn calculation?

You should NEVER intermingle free and paid users when calculating churn. They don't operate in similar manners. Track free user engagement on its own and concentrate on activation and conversion, not the usual churn numbers.

How do you predict which customers will churn?

The churn prediction of today uses data on how people use a service, how engaged they are and their demographics. The often-quoted claim that machine learning models can identify customers likely to leave 30 to 90 days in advance with 80 to 90 percent accuracy could not be found on the linked Pecan AI page, which warns that accuracy alone can mislead, because a model that predicts nobody leaves can still score high.

What does customer lifetime value have to do with churn rate?

LTV = Average Revenue Per User / Churn Rate, so the two are inversely related. ChartMogul gives LTV as average revenue per account divided by customer churn rate, which means halving churn doubles LTV; few things a subscription business can do have more upside than cutting churn.

How rapidly should you intervene on signs of churn?

Speed matters. Customers who say they are leaving have often made the decision weeks earlier. The often-quoted claim that most successful interventions occur within 24 to 48 hours of risk signals could not be found on the linked SuperOffice page.

Related Terms

Customer Lifetime Value (CLV) - Total revenue expected from customer relationship over entire duration
Customer Acquisition Cost (CAC) - Total cost invested to acquire a new customer through marketing efforts
Monthly Recurring Revenue (MRR) - Predictable revenue from subscriptions calculated monthly that churn directly impacts
Customer Retention - Strategies to keep existing customers engaged and loyal over time

Best practices and expert advice

i

"In today's competitive subscription economy, understanding churn isn't just about calculating numbers, it's about recognizing that each churned customer represents a failure to deliver ongoing value. The most successful companies I've worked with since 2008 treat churn reduction as a strategic imperative that touches every aspect of their customer experience."

Tessar Napitupulu, Founder and CEO of Arfadia

This is why so many startups hit growth walls, simple retention problems were unaddressed because the company was busy growing by the acquisition of new customers. What is the answer? Assign a named success rep to each of your customers.

The quote often attributed to ChurnZero CEO You Mon Tsang, that companies lose many customers on the way to product-market fit and should then look for another group of people who love the product, could not be found on the linked ChurnZero page or elsewhere.

Technology alone won't stop churn. The best retention methods combine advanced analytics with human judgment: predict which customers might leave, but when it's time to intervene, have real people reach out with tailored fixes.

Looking ahead: What will churn prevention look like?

There are a few trends shaping how companies retain customers:

Artificial intelligence-driven personalization is advancing past item recommendations to forging distinct experiences with users. That journey is unique to each customer because it is continually changing in the moment, based on that person's behavior.

Predictive analytics keeps improving. The most advanced models aim to tell you not only who is likely to leave, but why and when, and what is most likely to help.

Creating a community is becoming a potent retention method, although the often-quoted claim that active community members are 2-3 times more likely to stay could not be traced to a source.

Convergence is encouraging companies to bundle subscriptions. As cable companies discovered, bundles greatly reduce churn by raising the cost of switching.

Mostly, the dialogue is shifting from retaining customers to enabling their success. The best companies don't obsess about just retention metrics, they obsess about customer outcomes. Retention occurs organically when customers achieve their objectives.

So the path ahead is obvious: invest in understanding what your customers want to achieve, build systems to help them achieve it and watch as better retention transforms the way in which your business operates.