Churn Rate: How to Calculate It and What Counts as Healthy for Your Model

Churn rate is the percentage of customers, or of revenue, you lose in a period, and it is the single biggest driver of lifetime value in any recurring or repeat-purchase business. Calculating it is easy. Deciding what counts as healthy is not, because the right level depends on your margin, your acquisition cost and how long you can wait to recover it. This guide gives the formula, worked examples including annualizing, the split between voluntary and involuntary churn, and a method for deriving the maximum churn your own model can tolerate. It belongs to unit economics and budgeting and forecasting, builds on customer retention rate, and leads to revenue churn vs. logo churn and how to reduce CAC.

The formula

Churn rate = Customers lost during the period / Customers at the start of the period

Example: a company starts a month with 500 customers and loses 25. Churn = 25 / 500 = 5%. The retention rate is the complement, 95 percent. New customers acquired during the month are not in either figure.

A variation uses the average of the starting and ending customer counts as the denominator. Either is acceptable if applied consistently, but the start-of-period version is simpler and ties to cohorts.

Annualizing churn

Monthly churn compounds. At 5 percent a month, the share of a cohort still present after twelve months is 0.95^12 = 0.540, so annual churn is 1 - 0.540 = 46.0%. The shortcut of multiplying by twelve (60 percent) overstates it, and taking the monthly rate as the annual rate understates it by a large margin. Convert with powers, and always label the period.

Customer churn and revenue churn

Customer (logo) churn counts accounts. Revenue churn counts the revenue those accounts represented. If the customers you lose are small, revenue churn is lower than logo churn, and the reverse if you lose large ones. Track both, and see revenue churn vs. logo churn for a worked comparison.

Voluntary and involuntary churn

Not every lost customer decided to leave.

• Voluntary churn: the customer cancels or stops buying because of price, fit, a competitor or a changed need.
• Involuntary churn: the customer is lost because a payment fails, a card expires or an account is not renewed through an error.

Example: a subscription business has total monthly churn of 5 percent, of which 1.5 points are failed payments. Better payment retry rules, card updater services and reminder emails might recover a third to a half of those, which would cut total churn from 5.0 percent to about 4.3 percent (if 0.7 points are recovered). With contribution of $20 a month and a simple lifetime formula, LTV would rise from 20 / 0.05 = $400 to 20 / 0.043 = $465, an increase of 16 percent for a fix that does not touch the product. Separate the two types in reporting, because the fixes are different.

What counts as healthy for your model

There is no single healthy churn rate. Subscription boxes, consumer apps, small-business software and enterprise software differ widely, and a figure that is excellent in one can be ruinous in another. A more useful question is: what is the highest churn rate this business can tolerate? You can derive it.

Method 1: from the LTV to CAC target

Using the simple lifetime formula, LTV = monthly contribution / churn. If you want a ratio of at least R, then churn must be no higher than monthly contribution / (R × CAC).

Example: monthly contribution $20, CAC $150, target ratio 3.

Maximum churn = 20 / (3 × 150) = 20 / 450 = 4.4% a month. Above that, the model cannot meet the target. At 4.4 percent, the expected lifetime is 22.5 months and LTV is $450.

Method 2: from payback

Payback must occur before most customers have gone. Payback here is 150 / 20 = 7.5 months. With monthly churn of 4.4 percent, the share still present at 7.5 months is 0.956^7.5 = about 0.71, so roughly 29 percent of the cohort has left before payback. Whether that is acceptable depends on your cash position. A lower tolerance for churn follows from a longer payback.

Method 3: from your funding

If the business can fund only a six-month payback, compute the churn at which the cohort's cumulative contribution by month six equals CAC, and treat that as the ceiling. This turns churn from a metric into a constraint.

Churn by tenure

Early churn is usually higher than later churn. A customer who has stayed a year is more likely to stay another month than one who joined last week. This means that an average churn rate across all customers, which mixes the two, can mislead. Track churn by age band:

For a cohort of 1,000, the survivors after month 3 are 1,000 × 0.92^3 = 779. After a further nine months at 4 percent, 779 × 0.96^9 = 779 × 0.692 = 539. A flat 4 percent for twelve months would have left 1,000 × 0.96^12 = 613. Using the average churn overstates survival, because it understates the early losses. The fix for early churn is usually onboarding. The fix for later churn is usually value delivered and price.

How to find the causes

1. Segment churn by channel, plan, product, size, geography and tenure.
2. Ask departing customers why, with a short exit survey, and compare with their behavior.
3. Look at behavior before leaving: reduced usage, support contacts, late payments, order gaps.
4. Identify at-risk customers with simple rules before building models.
5. Test interventions with holdouts, and measure the saved customers.

Churn and operations

For product businesses, churn has operational causes: stockouts on the item a customer reorders, late deliveries, damaged goods and inconsistent quality. These often explain churn patterns that marketing cannot fix. See demand planning and inventory optimization.

Common mistakes

• Using a blended churn rate for segments with different behavior.
• Multiplying monthly churn by twelve.
• Counting new customers in the denominator.
• Ignoring involuntary churn.
• Reporting only logo churn when customers vary greatly in size.
• Comparing with benchmarks from businesses with different pricing, contracts or customers.
• Treating churn as a marketing problem only.

Churn in low-frequency consumer businesses

In businesses where customers buy a few times a year, a monthly churn rate is not meaningful, because customers who have not bought this month have not necessarily left. Use a window that matches the purchase cycle. If the typical repurchase interval is four months, define a customer as active if they bought in the last eight months, and measure churn as the share of active customers who become inactive over a quarter. Say plainly which definition you are using, and keep it constant. Without that discipline, a seasonal product will appear to have enormous churn every off season.

Setting up churn reporting

A useful monthly churn report has five parts: total churn, voluntary and involuntary churn, churn by tenure band, churn by segment, and a short explanation of the largest movements. Put the maximum tolerable churn, derived from your own model, as a line on the chart so that readers can see at a glance how far actual churn sits from the limit. Review the report with product, marketing and customer service together, since each controls part of the cause.

FAQs

What is churn rate?

The percentage of customers or revenue lost during a period, calculated as the number lost divided by the number at the start.

How do you calculate churn rate?

Divide customers lost during the period by customers at the start of the period. Exclude new customers acquired during the period.

What is a good churn rate?

It depends on your model. Calculate the maximum churn your LTV to CAC target and payback can tolerate, and compare your actual churn with that limit.

How do you convert monthly churn to annual churn?

Subtract the monthly retention rate raised to the twelfth power from one. A monthly churn of 5 percent equals annual churn of about 46 percent.

What is the difference between voluntary and involuntary churn?

Voluntary churn is a customer's decision to leave. Involuntary churn comes from failed payments or administrative errors.