Customer Churn: Definition, Rate Formula, and How to Reduce It

Customer churn is the loss of customers who stop buying from you or cancel what they pay for. It is one of the clearest signs that your product, price, or service is falling short.

Most teams notice churn only after the revenue has gone. A better approach is to measure it cleanly, learn why people leave, and spot the warning signs early.

In this guide, you will learn how to calculate your customer churn rate and read a churn analysis. You will also see how to predict who may leave next and reduce churn with six practical tactics.

What is customer churn?

Customer churn is the percentage of customers who stop doing business with a company during a set period. It is also called customer attrition. A customer has churned once they cancel, let a contract lapse, or stop buying altogether.

Churn matters because keeping customers costs far less than replacing them. Harvard Business Review reports that acquiring a new customer can cost 5 to 25 times more than retaining an existing one. It also cites Bain & Company research showing that a 5% rise in retention can lift profits by 25% to 95%.

A customer counts as churned in different ways, depending on your model:

• Subscription: The customer cancels, or the plan expires.
• Contract: The customer does not renew at term end.
• Transactional: The customer passes an inactivity window with no purchase.

Customer churn vs. retention, attrition, and revenue churn

Several terms get used as if they mean the same thing. They do not, and mixing them up leads to wrong calculations.

A business can lose 5% of its customers and only 2% of its revenue if the leavers were small accounts. The reverse can also happen. That is why you should track both numbers.

Retention is the mirror image of churn. This guide to customer retention rate covers that side in full, including how to calculate it correctly.

How do you calculate customer churn rate?

Divide the customers you lost during a period by the customers you had at the start, then multiply by 100. Pick one period, such as a month or a quarter, and use it every time.

1. Choose your time period (monthly, quarterly, or annual).
2. Count the customers you had on the first day of that period.
3. Count how many of those same customers left by the last day.
4. Divide lost customers by starting customers, then multiply by 100.

Worked example: A software company starts the quarter with 2,000 customers and loses 90 of them. The churn rate is 90 ÷ 2,000 × 100, which equals 4.5%. The retention rate for the same group is 95.5%.

New customers who joined mid-quarter stay out of the math. They were not in the starting group, so counting them would hide real losses.

How do you calculate revenue churn?

Revenue churn works the same way, but it counts money instead of people.

• Formula: Revenue lost from existing customers ÷ revenue at the start × 100
• Example: You begin the month with $200,000 in monthly recurring revenue (MRR, the predictable income from subscriptions) and lose $14,000. Revenue churn is 7%.

What is a good customer churn rate?

A good churn rate depends on your industry, customer size, and billing model. Lower is always better, but the right target varies widely.

Recurly’s 2026 churn rate benchmarks give a useful reference point:

• Average voluntary churn is 2.34% and average involuntary churn is 1.25%.
• Voluntary churn makes up the majority of total churn in every industry Recurly tracks.
• Ecommerce shows total churn of 4.25%, while business and professional services show 3.44%.

Use these figures as a starting reference, not a verdict. Compare your numbers over the same time period, and watch your own trend. A steady fall over six months tells you more than a single benchmark match.

Why small monthly numbers add up

Monthly churn compounds across a year, and it also shortens how long a customer stays. Divide 1 by the monthly churn rate to estimate average customer lifetime in months.

• 2% monthly churn: You lose about 22% of customers a year, and the average customer stays 50 months.
• 3% monthly churn: About 31% a year, with a 33-month average stay.
• 5% monthly churn: About 46% a year, with a 20-month average stay.
• 8% monthly churn: About 63% a year, with a 12.5-month average stay.

What are the types of customer churn?

Churn falls into a few types, and each one needs a different fix. Naming the type you face saves months of wasted effort.

Gradual churn deserves extra attention. It often looks like healthy retention on paper, yet it signals a full cancellation to come.

What causes customers to churn?

Customers leave when the value they get no longer matches the price or effort. The reasons usually fall into a handful of patterns.

• Poor onboarding: New customers never reach a first win and drift away.
• Weak support: Slow or unhelpful service after a problem leaves a lasting mark.
• High effort: Tasks that feel harder than they should push people toward easier options.
• Price mismatch: A competitor offers similar value for less, or a price rise lands badly.
• Product fit: The product never solved the original problem.
• Silence: No one checks in until the renewal date arrives.

Many of these show up in feedback long before the cancellation. That is why listening programs, often called voice of the customer, matter. This voice of customer analytics guide explains how to collect and read that feedback.

Customer churn examples

Real situations make the numbers easier to read. One example below uses public reporting. The other three are illustrative scenarios, not company data.

Netflix and the 2011 price change

In 2011, Netflix raised prices and split its DVD and streaming plans. According to CleverTap’s churn analysis, the company lost about 4% of its U.S. subscribers and roughly $8 million in monthly revenue.

• Cause: a price change customers saw as unfair.
• Lesson: price shifts need clear value messaging, and feedback should be tracked before and after.

A subscription software company (illustrative)

A project management tool has 1,500 customers. Fifty cancel in one month, a 3.3% monthly churn rate. Exit surveys show 60% of leavers never set up integrations.

The fix is a guided onboarding flow, not a discount. The data pointed to a product-adoption gap, not a price problem.

An online retailer (illustrative)

An ecommerce store counts a buyer as churned after 180 days without a purchase.

That 28% rate prompts the team to launch a win-back email series and ask lapsed buyers one short question about why they stopped.

A regional fitness chain (illustrative)

A gym group sees most cancellations in months two and three after a sign-up promotion. Cohort data points to a weak first-month experience.

The chain adds a free coaching session in week two. Early cancellations drop because members now have a reason to keep showing up.

How do you run a customer churn analysis?

A customer churn analysis finds who is leaving, when, and why. The goal is to move from one overall percentage to specific causes you can fix.

Three methods cover most needs:

• Cohort analysis.
  Group customers by sign-up month and compare how long each group stays.

• Segment analysis.
  Split churn by plan, region, industry, or acquisition channel. One weak segment can hide behind a healthy average.

• Behavior analysis.
  Compare the usage patterns of leavers and stayers. Features that stayers use and leavers skip show you where value lives.

What does a cohort table look like?

Each row is a sign-up group, and each column shows the share still active. These figures are illustrative.

The March group loses customers faster. Something changed between January and March, such as a new campaign, a pricing change, or an onboarding gap. That gap is where to dig next.

Pair these numbers with direct feedback. Data tells you what happened, and customers tell you why.

How can surveys predict customer churn?

Surveys predict churn by catching dissatisfaction while a customer is still active. Three metrics do most of the work.

• Net Promoter Score (NPS): Asks how likely a customer is to recommend you on a 0 to 10 scale. Scores of 0 to 6 mark detractors, your highest churn risk.
• Customer Effort Score (CES): Measures how easy it was to get something done. Rising effort after a support or billing contact is a warning sign.
• Customer satisfaction (CSAT): Captures satisfaction after one interaction. Repeated low scores from the same account deserve a call.

Good question design makes these scores more useful. These customer satisfaction survey questions show how to phrase them so the answers lead to action.

Set a rule for each score. For example, route every detractor response to an account owner within 48 hours.

How does churn prediction work?

Churn prediction scores each customer on the odds they will leave, using signals that tend to appear before a cancellation. A simple version needs no data science team.

1. Pick four to six signals.
   Common ones include a usage drop, a failed payment, a low NPS answer, an open support ticket, and no login for 30 days.

2. Weight each signal.
   A failed payment and a detractor score together should rank higher than a single quiet week.

3. Set thresholds.
   For example, flag any account with a 20% month-over-month usage drop. Tune the number to your own data.

4. Assign an owner.
   Each flagged account needs one person responsible for a follow-up.

5. Review monthly.
   Compare flagged accounts with real cancellations, then adjust the weights.

Machine learning can refine these scores over time. Start with a rule-based model first, because it is easy to explain and quick to improve.

How do you decide which churn to fix first?

Fix the churn that costs the most revenue and has the clearest cause. Lost customer counts alone can mislead you, because small and large accounts leave at different rates.

Take two segments. One group loses 40 customers at $50 a month each, or $2,000 in monthly revenue. Another loses 10 customers at $400 a month each, or $4,000.

That second group loses fewer customers but twice the revenue. Use this order of work:

• Rank by revenue lost: Use revenue at risk, not head count.
• Sort by cause: Separate fixable causes (failed payments, weak onboarding) from structural ones (poor product fit).
• Estimate effort: Score each fix as low, medium, or high effort.
• Pilot first: Start with high-revenue, low-effort fixes on one segment.
• Compare results: Measure churn before and after for that segment alone.

Also check customer lifetime value, the total revenue you expect from a customer over the relationship. Losing a customer with high lifetime value costs more than the monthly figure suggests.

How to reduce customer churn: 6 Practical ways

Reducing churn means fixing the causes you found, one at a time. These six tactics work across most industries.

1. Fix onboarding first

The first 30 days set the tone. Define the one action that proves value, and guide every new customer to it quickly.

• Send a welcome message that names the first step.
• Check progress at day 7 and day 21.

2. Ask why customers leave

Run a short exit survey at cancellation. Two or three questions are enough.

• What was the main reason you left?
• What could we have done differently?

Group the answers into themes every month.

3. Act on low scores fast

A detractor response is a flare. Follow up personally, solve the issue, and record the outcome. This practice is called closed-loop feedback, and it turns a complaint into a second chance.

• Reply within 48 hours.
• Log what you fixed, so patterns show up.

4. Reward loyal customers

Perks, early access, and fair renewal pricing show appreciation. Track results with these customer loyalty metrics to see whether your program changes behavior.

• Reward tenure, not just spend.
• Offer renewal pricing before the customer asks.

5. Reduce customer effort

Remove steps from sign-up, billing, and support. Every extra click gives a customer a reason to look elsewhere.

• Audit your top three support requests.
• Shorten the cancellation-to-help path, so you can offer a fix first.

6. Fix failed payments

For involuntary churn, send card-expiry reminders and retry failed charges at sensible intervals. This often recovers revenue that never needed to be lost.

• Email customers 14 days before a card expires.
• Retry failed charges on a set schedule.

What mistakes make churn data misleading?

Bad measurement leads to bad decisions. Each error below hides a real problem or invents a false one.

A clean definition of “churned” that every team shares prevents most of these. For a wider set of tracking options, see this list of customer retention metrics.

How does QuestionPro Customer Experience help spot churn risk?

QuestionPro Customer Experience puts NPS, CSAT, and CES in one place and flags churn risk from customer sentiment. Two capabilities matter most for churn work:

• Closed-loop case management: Low scores route to the right owner, so a detractor gets a reply instead of silence.
• AskWhy: One question captures the score, the reason, and the comment together, which gives you the “why” without a long survey.

Text analytics then groups open-ended answers into themes. You can see which causes drive the most cancellations and act on the biggest one first.

Churn is a symptom, not the problem

Churn tells you that a customer stopped finding enough value. The percentage fixes nothing on its own. What matters is the quality of the questions you ask once you see it.

Teams that cut churn do three things well:

• Measure cleanly: One definition, one period, both customer and revenue views.
• Listen early: Surveys and signals before the cancellation form appears.
• Close the loop: A real reply when someone is unhappy.

Start with one segment, one survey, and one fix, then repeat.

Frequently Asked Questions (FAQs)